diff --git a/.gitattributes b/.gitattributes index ecaef5ea65967aa14f9963a41681b8c1024ea593..5727c7b07e343cbb00e7b69fd26fcd6ede46a91d 100644 --- a/.gitattributes +++ b/.gitattributes @@ -10618,3 +10618,24 @@ neuronxcc-2.19.8089.0+8ab9f450/MODULE_0bac481d29ac1829a8af+253d6470/model.neff f neuronxcc-2.19.8089.0+8ab9f450/MODULE_2052cf656c210489741d+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text neuronxcc-2.19.8089.0+8ab9f450/MODULE_b9a3b941bc75ca360b60+cd3419b6/model.neff filter=lfs diff=lfs merge=lfs -text neuronxcc-2.19.8089.0+8ab9f450/MODULE_b9a3b941bc75ca360b60+cd3419b6/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text +neuronxcc-2.19.8089.0+8ab9f450/MODULE_0a47deab436eaf26c99a+cd3419b6/model.neff filter=lfs diff=lfs merge=lfs -text +neuronxcc-2.19.8089.0+8ab9f450/MODULE_0a47deab436eaf26c99a+cd3419b6/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text +neuronxcc-2.19.8089.0+8ab9f450/MODULE_0b67a734fc55d67768bd+253d6470/model.neff filter=lfs diff=lfs merge=lfs -text +neuronxcc-2.19.8089.0+8ab9f450/MODULE_184fcc045921c74845be+cd3419b6/model.neff filter=lfs diff=lfs merge=lfs -text +neuronxcc-2.19.8089.0+8ab9f450/MODULE_184fcc045921c74845be+cd3419b6/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text +neuronxcc-2.19.8089.0+8ab9f450/MODULE_1d0c9d2357e7888e46cd+253d6470/model.neff filter=lfs diff=lfs merge=lfs -text +neuronxcc-2.19.8089.0+8ab9f450/MODULE_20ce63f489314924a057+253d6470/model.neff filter=lfs diff=lfs merge=lfs -text +neuronxcc-2.19.8089.0+8ab9f450/MODULE_40f6b4304be01d72d956+253d6470/model.neff filter=lfs diff=lfs merge=lfs -text +neuronxcc-2.19.8089.0+8ab9f450/MODULE_64f5633fb25cfe2548d6+253d6470/model.neff filter=lfs diff=lfs merge=lfs -text +neuronxcc-2.19.8089.0+8ab9f450/MODULE_806ef7cec3a07d99b574+cd3419b6/model.neff filter=lfs diff=lfs merge=lfs -text +neuronxcc-2.19.8089.0+8ab9f450/MODULE_806ef7cec3a07d99b574+cd3419b6/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text +neuronxcc-2.19.8089.0+8ab9f450/MODULE_9f17b2a2f6b76b4ae4a8+253d6470/model.neff filter=lfs diff=lfs merge=lfs -text +neuronxcc-2.19.8089.0+8ab9f450/MODULE_b522f7aabb6b091bde5f+253d6470/model.neff filter=lfs diff=lfs merge=lfs -text +neuronxcc-2.19.8089.0+8ab9f450/MODULE_b72befbd04b196887e2e+cd3419b6/model.neff filter=lfs diff=lfs merge=lfs -text +neuronxcc-2.19.8089.0+8ab9f450/MODULE_b72befbd04b196887e2e+cd3419b6/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text +neuronxcc-2.19.8089.0+8ab9f450/MODULE_c5daaeaa40996f359a5c+cd3419b6/model.neff filter=lfs diff=lfs merge=lfs -text +neuronxcc-2.19.8089.0+8ab9f450/MODULE_c5daaeaa40996f359a5c+cd3419b6/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text +neuronxcc-2.19.8089.0+8ab9f450/MODULE_df21c2b35cb01f251b78+cd3419b6/model.neff filter=lfs diff=lfs merge=lfs -text +neuronxcc-2.19.8089.0+8ab9f450/MODULE_df21c2b35cb01f251b78+cd3419b6/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text +neuronxcc-2.19.8089.0+8ab9f450/MODULE_f561bf7d12efdc67186d+cd3419b6/model.neff filter=lfs diff=lfs merge=lfs -text +neuronxcc-2.19.8089.0+8ab9f450/MODULE_f561bf7d12efdc67186d+cd3419b6/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/Qwen/Qwen3-30B-A3B-Instruct-2507/2e075469bb7e246e19de.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/Qwen/Qwen3-30B-A3B-Instruct-2507/2e075469bb7e246e19de.json new file mode 100644 index 0000000000000000000000000000000000000000..4d96cdd82a33c2bf0e3e35ce5dbeeea2cafd45f9 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/Qwen/Qwen3-30B-A3B-Instruct-2507/2e075469bb7e246e19de.json @@ -0,0 +1,79 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "Qwen/Qwen3-30B-A3B-Instruct-2507", + "_task": "text-generation", + "architectures": [ + "Qwen3MoeForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "decoder_sparse_step": 1, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 2048, + "initializer_range": 0.02, + "intermediate_size": 6144, + "max_position_embeddings": 262144, + "max_window_layers": 48, + "mlp_only_layers": [], + "model_type": "qwen3_moe", + "moe_intermediate_size": 768, + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 8, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "Qwen/Qwen3-30B-A3B-Instruct-2507", + "checkpoint_revision": "61082d4deaa4785f64943b443cbc2b5de7524fad", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 8, + "logical_nc_config": 1, + "max_batch_size": 8, + "max_context_length": 4096, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 4096, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": false, + "optimum_neuron_version": "0.3.1.dev1", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "bfloat16", + "sequence_length": 4096, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "bfloat16", + "tp_degree": 8, + "vocab_parallel": false + }, + "norm_topk_prob": true, + "num_attention_heads": 32, + "num_experts": 128, + "num_experts_per_tok": 8, + "num_hidden_layers": 48, + "num_key_value_heads": 4, + "output_router_logits": false, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 10000000, + "router_aux_loss_coef": 0.001, + "sliding_window": null, + "tie_word_embeddings": false, + "use_cache": true, + "use_sliding_window": false, + "vocab_size": 151936 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/63936687f0615be0c4b5.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/63936687f0615be0c4b5.json new file mode 100644 index 0000000000000000000000000000000000000000..c31686cfbc1c1a9d54c0d86dcb4bded5dca641cb --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/63936687f0615be0c4b5.json @@ -0,0 +1,79 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "optimum-internal-testing/tiny-random-qwen3_moe", + "_task": "text-generation", + "architectures": [ + "Qwen3MoeForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "decoder_sparse_step": 2, + "head_dim": 32, + "hidden_act": "silu", + "hidden_size": 64, + "initializer_range": 0.02, + "intermediate_size": 128, + "max_position_embeddings": 40960, + "max_window_layers": 1, + "mlp_only_layers": [], + "model_type": "qwen3_moe", + "moe_intermediate_size": 128, + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 1, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "optimum-internal-testing/tiny-random-qwen3_moe", + "checkpoint_revision": "e0230be2839556b44b7400a233c73c74b4abb7af", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 1, + "max_context_length": 1024, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 1024, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": false, + "optimum_neuron_version": "0.3.1.dev1", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "bfloat16", + "sequence_length": 1024, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "bfloat16", + "tp_degree": 2, + "vocab_parallel": false + }, + "norm_topk_prob": true, + "num_attention_heads": 2, + "num_experts": 8, + "num_experts_per_tok": 2, + "num_hidden_layers": 2, + "num_key_value_heads": 1, + "output_router_logits": false, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 1000000.0, + "router_aux_loss_coef": 0.001, + "sliding_window": null, + "tie_word_embeddings": true, + "use_cache": true, + "use_sliding_window": false, + "vocab_size": 151936 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/67d11d4985c28ac7d5ff.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/67d11d4985c28ac7d5ff.json new file mode 100644 index 0000000000000000000000000000000000000000..5e92538d4c65e2230433c39b14581d471147e741 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/67d11d4985c28ac7d5ff.json @@ -0,0 +1,79 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "optimum-internal-testing/tiny-random-qwen3_moe", + "_task": "text-generation", + "architectures": [ + "Qwen3MoeForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "decoder_sparse_step": 2, + "head_dim": 32, + "hidden_act": "silu", + "hidden_size": 64, + "initializer_range": 0.02, + "intermediate_size": 128, + "max_position_embeddings": 40960, + "max_window_layers": 1, + "mlp_only_layers": [], + "model_type": "qwen3_moe", + "moe_intermediate_size": 128, + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 1, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "optimum-internal-testing/tiny-random-qwen3_moe", + "checkpoint_revision": "e0230be2839556b44b7400a233c73c74b4abb7af", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 1, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": false, + "optimum_neuron_version": "0.3.1.dev1", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "bfloat16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "bfloat16", + "tp_degree": 2, + "vocab_parallel": false + }, + "norm_topk_prob": true, + "num_attention_heads": 2, + "num_experts": 8, + "num_experts_per_tok": 2, + "num_hidden_layers": 2, + "num_key_value_heads": 1, + "output_router_logits": false, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 1000000.0, + "router_aux_loss_coef": 0.001, + "sliding_window": null, + "tie_word_embeddings": true, + "use_cache": true, + "use_sliding_window": false, + "vocab_size": 151936 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/cd03fa772b07fbb454c0.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/cd03fa772b07fbb454c0.json new file mode 100644 index 0000000000000000000000000000000000000000..857271cc14484c5f4141af1f56bde95321ddbf7d --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/cd03fa772b07fbb454c0.json @@ -0,0 +1,79 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "optimum-internal-testing/tiny-random-qwen3_moe", + "_task": "text-generation", + "architectures": [ + "Qwen3MoeForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "decoder_sparse_step": 2, + "head_dim": 32, + "hidden_act": "silu", + "hidden_size": 64, + "initializer_range": 0.02, + "intermediate_size": 128, + "max_position_embeddings": 40960, + "max_window_layers": 1, + "mlp_only_layers": [], + "model_type": "qwen3_moe", + "moe_intermediate_size": 128, + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 1, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "optimum-internal-testing/tiny-random-qwen3_moe", + "checkpoint_revision": "e0230be2839556b44b7400a233c73c74b4abb7af", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 1, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": false, + "optimum_neuron_version": "0.3.1.dev1", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "float16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "float16", + "tp_degree": 2, + "vocab_parallel": false + }, + "norm_topk_prob": true, + "num_attention_heads": 2, + "num_experts": 8, + "num_experts_per_tok": 2, + "num_hidden_layers": 2, + "num_key_value_heads": 1, + "output_router_logits": false, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 1000000.0, + "router_aux_loss_coef": 0.001, + "sliding_window": null, + "tie_word_embeddings": true, + "use_cache": true, + "use_sliding_window": false, + "vocab_size": 151936 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/e5eddc28add59e2ee9d6.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/e5eddc28add59e2ee9d6.json new file mode 100644 index 0000000000000000000000000000000000000000..9861451f01a555e63d7d11f1c7c18529098a1ee1 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/e5eddc28add59e2ee9d6.json @@ -0,0 +1,79 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "optimum-internal-testing/tiny-random-qwen3_moe", + "_task": "text-generation", + "architectures": [ + "Qwen3MoeForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "decoder_sparse_step": 2, + "head_dim": 32, + "hidden_act": "silu", + "hidden_size": 64, + "initializer_range": 0.02, + "intermediate_size": 128, + "max_position_embeddings": 40960, + "max_window_layers": 1, + "mlp_only_layers": [], + "model_type": "qwen3_moe", + "moe_intermediate_size": 128, + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 2, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "optimum-internal-testing/tiny-random-qwen3_moe", + "checkpoint_revision": "e0230be2839556b44b7400a233c73c74b4abb7af", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 2, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": false, + "optimum_neuron_version": "0.3.1.dev1", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "float16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "float16", + "tp_degree": 2, + "vocab_parallel": false + }, + "norm_topk_prob": true, + "num_attention_heads": 2, + "num_experts": 8, + "num_experts_per_tok": 2, + "num_hidden_layers": 2, + "num_key_value_heads": 1, + "output_router_logits": false, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 1000000.0, + "router_aux_loss_coef": 0.001, + "sliding_window": null, + "tie_word_embeddings": true, + "use_cache": true, + "use_sliding_window": false, + "vocab_size": 151936 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/tiny-random/qwen3-moe/21c419fc8bcc3104cc37.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/tiny-random/qwen3-moe/21c419fc8bcc3104cc37.json new file mode 100644 index 0000000000000000000000000000000000000000..f42327d2bbeeb9b31bab20146edd8583d89f4b9b --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/tiny-random/qwen3-moe/21c419fc8bcc3104cc37.json @@ -0,0 +1,79 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "tiny-random/qwen3-moe", + "_task": "text-generation", + "architectures": [ + "Qwen3MoeForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "decoder_sparse_step": 2, + "head_dim": 32, + "hidden_act": "silu", + "hidden_size": 64, + "initializer_range": 0.02, + "intermediate_size": 128, + "max_position_embeddings": 40960, + "max_window_layers": 1, + "mlp_only_layers": [], + "model_type": "qwen3_moe", + "moe_intermediate_size": 128, + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 2, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "tiny-random/qwen3-moe", + "checkpoint_revision": "10a349dcb488b10c27aa4a3c1dbefb74c41565c3", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 2, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": false, + "optimum_neuron_version": "0.3.1.dev1", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "float16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "float16", + "tp_degree": 2, + "vocab_parallel": false + }, + "norm_topk_prob": true, + "num_attention_heads": 2, + "num_experts": 8, + "num_experts_per_tok": 2, + "num_hidden_layers": 2, + "num_key_value_heads": 1, + "output_router_logits": false, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 1000000.0, + "router_aux_loss_coef": 0.001, + "sliding_window": null, + "tie_word_embeddings": true, + "use_cache": true, + "use_sliding_window": false, + "vocab_size": 151936 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/tiny-random/qwen3-moe/e14cd546176a864ee18b.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/tiny-random/qwen3-moe/e14cd546176a864ee18b.json new file mode 100644 index 0000000000000000000000000000000000000000..4a14615eca9f58adc68a1a996b8bc6935a75dc92 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/tiny-random/qwen3-moe/e14cd546176a864ee18b.json @@ -0,0 +1,79 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "tiny-random/qwen3-moe", + "_task": "text-generation", + "architectures": [ + "Qwen3MoeForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "decoder_sparse_step": 2, + "head_dim": 32, + "hidden_act": "silu", + "hidden_size": 64, + "initializer_range": 0.02, + "intermediate_size": 128, + "max_position_embeddings": 40960, + "max_window_layers": 1, + "mlp_only_layers": [], + "model_type": "qwen3_moe", + "moe_intermediate_size": 128, + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 1, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "tiny-random/qwen3-moe", + "checkpoint_revision": "10a349dcb488b10c27aa4a3c1dbefb74c41565c3", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 1, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": false, + "optimum_neuron_version": "0.3.1.dev1", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "bfloat16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "bfloat16", + "tp_degree": 2, + "vocab_parallel": false + }, + "norm_topk_prob": true, + "num_attention_heads": 2, + "num_experts": 8, + "num_experts_per_tok": 2, + "num_hidden_layers": 2, + "num_key_value_heads": 1, + "output_router_logits": false, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 1000000.0, + "router_aux_loss_coef": 0.001, + "sliding_window": null, + "tie_word_embeddings": true, + "use_cache": true, + "use_sliding_window": false, + "vocab_size": 151936 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/tiny-random/qwen3-moe/e99ce6fcf5be305e9424.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/tiny-random/qwen3-moe/e99ce6fcf5be305e9424.json new file mode 100644 index 0000000000000000000000000000000000000000..fe4648571c25856e0515e4f7d8dcbeb9f4912c40 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/tiny-random/qwen3-moe/e99ce6fcf5be305e9424.json @@ -0,0 +1,79 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "tiny-random/qwen3-moe", + "_task": "text-generation", + "architectures": [ + "Qwen3MoeForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "decoder_sparse_step": 2, + "head_dim": 32, + "hidden_act": "silu", + "hidden_size": 64, + "initializer_range": 0.02, + "intermediate_size": 128, + "max_position_embeddings": 40960, + "max_window_layers": 1, + "mlp_only_layers": [], + "model_type": "qwen3_moe", + "moe_intermediate_size": 128, + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 1, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "tiny-random/qwen3-moe", + "checkpoint_revision": "10a349dcb488b10c27aa4a3c1dbefb74c41565c3", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 1, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": false, + "optimum_neuron_version": "0.3.1.dev1", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "float16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "float16", + "tp_degree": 2, + "vocab_parallel": false + }, + "norm_topk_prob": true, + "num_attention_heads": 2, + "num_experts": 8, + "num_experts_per_tok": 2, + "num_hidden_layers": 2, + "num_key_value_heads": 1, + "output_router_logits": false, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 1000000.0, + "router_aux_loss_coef": 0.001, + "sliding_window": null, + "tie_word_embeddings": true, + "use_cache": true, + "use_sliding_window": false, + "vocab_size": 151936 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/yujiepan/qwen3-moe-tiny-random/1cacc2cc42865a8b000d.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/yujiepan/qwen3-moe-tiny-random/1cacc2cc42865a8b000d.json new file mode 100644 index 0000000000000000000000000000000000000000..83443d345a27a77ac798d282e6139485f26ec536 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/yujiepan/qwen3-moe-tiny-random/1cacc2cc42865a8b000d.json @@ -0,0 +1,79 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "yujiepan/qwen3-moe-tiny-random", + "_task": "text-generation", + "architectures": [ + "Qwen3MoeForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "decoder_sparse_step": 2, + "head_dim": 32, + "hidden_act": "silu", + "hidden_size": 64, + "initializer_range": 0.02, + "intermediate_size": 128, + "max_position_embeddings": 40960, + "max_window_layers": 1, + "mlp_only_layers": [], + "model_type": "qwen3_moe", + "moe_intermediate_size": 128, + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 2, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "yujiepan/qwen3-moe-tiny-random", + "checkpoint_revision": "fb6c5ee2a2c19bd9aced6d9afd8a858966a7bb7e", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 2, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": false, + "optimum_neuron_version": "0.3.1.dev1", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "float16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "float16", + "tp_degree": 2, + "vocab_parallel": false + }, + "norm_topk_prob": true, + "num_attention_heads": 2, + "num_experts": 8, + "num_experts_per_tok": 2, + "num_hidden_layers": 2, + "num_key_value_heads": 1, + "output_router_logits": false, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 1000000.0, + "router_aux_loss_coef": 0.001, + "sliding_window": null, + "tie_word_embeddings": true, + "use_cache": true, + "use_sliding_window": false, + "vocab_size": 151936 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/yujiepan/qwen3-moe-tiny-random/619f2f5356bdf0f59205.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/yujiepan/qwen3-moe-tiny-random/619f2f5356bdf0f59205.json new file mode 100644 index 0000000000000000000000000000000000000000..b29c4d5d735011faac8c3e32b97562f02e07fb4a --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/yujiepan/qwen3-moe-tiny-random/619f2f5356bdf0f59205.json @@ -0,0 +1,79 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "yujiepan/qwen3-moe-tiny-random", + "_task": "text-generation", + "architectures": [ + "Qwen3MoeForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "decoder_sparse_step": 2, + "head_dim": 32, + "hidden_act": "silu", + "hidden_size": 64, + "initializer_range": 0.02, + "intermediate_size": 128, + "max_position_embeddings": 40960, + "max_window_layers": 1, + "mlp_only_layers": [], + "model_type": "qwen3_moe", + "moe_intermediate_size": 128, + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 1, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "yujiepan/qwen3-moe-tiny-random", + "checkpoint_revision": "fb6c5ee2a2c19bd9aced6d9afd8a858966a7bb7e", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 1, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": false, + "optimum_neuron_version": "0.3.1.dev1", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "bfloat16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "bfloat16", + "tp_degree": 2, + "vocab_parallel": false + }, + "norm_topk_prob": true, + "num_attention_heads": 2, + "num_experts": 8, + "num_experts_per_tok": 2, + "num_hidden_layers": 2, + "num_key_value_heads": 1, + "output_router_logits": false, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 1000000.0, + "router_aux_loss_coef": 0.001, + "sliding_window": null, + "tie_word_embeddings": true, + "use_cache": true, + "use_sliding_window": false, + "vocab_size": 151936 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/yujiepan/qwen3-moe-tiny-random/67bc8c12b9b1221b38ad.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/yujiepan/qwen3-moe-tiny-random/67bc8c12b9b1221b38ad.json new file mode 100644 index 0000000000000000000000000000000000000000..b9c74ffad3ea40cbac21d589c5d0de91e56298e8 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/qwen3_moe/yujiepan/qwen3-moe-tiny-random/67bc8c12b9b1221b38ad.json @@ -0,0 +1,79 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "yujiepan/qwen3-moe-tiny-random", + "_task": "text-generation", + "architectures": [ + "Qwen3MoeForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "decoder_sparse_step": 2, + "head_dim": 32, + "hidden_act": "silu", + "hidden_size": 64, + "initializer_range": 0.02, + "intermediate_size": 128, + "max_position_embeddings": 40960, + "max_window_layers": 1, + "mlp_only_layers": [], + "model_type": "qwen3_moe", + "moe_intermediate_size": 128, + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 1, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "yujiepan/qwen3-moe-tiny-random", + "checkpoint_revision": "fb6c5ee2a2c19bd9aced6d9afd8a858966a7bb7e", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 1, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": false, + "optimum_neuron_version": "0.3.1.dev1", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "float16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "float16", + "tp_degree": 2, + "vocab_parallel": false + }, + "norm_topk_prob": true, + "num_attention_heads": 2, + "num_experts": 8, + "num_experts_per_tok": 2, + "num_hidden_layers": 2, + "num_key_value_heads": 1, + "output_router_logits": false, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 1000000.0, + "router_aux_loss_coef": 0.001, + "sliding_window": null, + "tie_word_embeddings": true, + "use_cache": true, + "use_sliding_window": false, + "vocab_size": 151936 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/0f44022ffda5d90427b3.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/0f44022ffda5d90427b3.json new file mode 100644 index 0000000000000000000000000000000000000000..1935bb29c34c11bee37f95305cb7d0d5231bd2df --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/0f44022ffda5d90427b3.json @@ -0,0 +1,72 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "hf-internal-testing/tiny-random-GraniteForCausalLM", + "_task": "text-generation", + "architectures": [ + "GraniteForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "attention_multiplier": 1.0, + "embedding_multiplier": 1.0, + "hidden_act": "silu", + "hidden_size": 32, + "initializer_range": 0.02, + "intermediate_size": 64, + "logits_scaling": 1.0, + "max_position_embeddings": 2048, + "mlp_bias": false, + "model_type": "granite", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 1, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "hf-internal-testing/tiny-random-GraniteForCausalLM", + "checkpoint_revision": "c3074ebc0ac2fe545305f5e5f6cce2cc9b2aa0c5", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": true, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 1, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": true, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "bfloat16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "bfloat16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 4, + "num_hidden_layers": 2, + "num_key_value_heads": 4, + "residual_multiplier": 1.0, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 10000.0, + "tie_word_embeddings": false, + "use_cache": true, + "vocab_size": 49152 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/1471022e22b5b8b3de3b.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/1471022e22b5b8b3de3b.json new file mode 100644 index 0000000000000000000000000000000000000000..d428d7f4b1e6cdbb54feb5b6ba487b3660c44aa9 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/1471022e22b5b8b3de3b.json @@ -0,0 +1,72 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "hf-internal-testing/tiny-random-GraniteForCausalLM", + "_task": "text-generation", + "architectures": [ + "GraniteForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "attention_multiplier": 1.0, + "embedding_multiplier": 1.0, + "hidden_act": "silu", + "hidden_size": 32, + "initializer_range": 0.02, + "intermediate_size": 64, + "logits_scaling": 1.0, + "max_position_embeddings": 2048, + "mlp_bias": false, + "model_type": "granite", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 1, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "hf-internal-testing/tiny-random-GraniteForCausalLM", + "checkpoint_revision": "c3074ebc0ac2fe545305f5e5f6cce2cc9b2aa0c5", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": true, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 1, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": true, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "float16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "float16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 4, + "num_hidden_layers": 2, + "num_key_value_heads": 4, + "residual_multiplier": 1.0, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 10000.0, + "tie_word_embeddings": false, + "use_cache": true, + "vocab_size": 49152 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/c9c911d79dbddb873dcf.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/c9c911d79dbddb873dcf.json new file mode 100644 index 0000000000000000000000000000000000000000..c4ffaddf9d60d04ccb46d0631af3014305b2c2f1 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/c9c911d79dbddb873dcf.json @@ -0,0 +1,72 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "hf-internal-testing/tiny-random-GraniteForCausalLM", + "_task": "text-generation", + "architectures": [ + "GraniteForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "attention_multiplier": 1.0, + "embedding_multiplier": 1.0, + "hidden_act": "silu", + "hidden_size": 32, + "initializer_range": 0.02, + "intermediate_size": 64, + "logits_scaling": 1.0, + "max_position_embeddings": 2048, + "mlp_bias": false, + "model_type": "granite", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 2, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "hf-internal-testing/tiny-random-GraniteForCausalLM", + "checkpoint_revision": "c3074ebc0ac2fe545305f5e5f6cce2cc9b2aa0c5", + "continuous_batching": true, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": true, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 2, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": true, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "float16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "float16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 4, + "num_hidden_layers": 2, + "num_key_value_heads": 4, + "residual_multiplier": 1.0, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 10000.0, + "tie_word_embeddings": false, + "use_cache": true, + "vocab_size": 49152 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/granite/ibm-granite/granite-3.1-2b-instruct/b66dbd045f3e73eb7427.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/granite/ibm-granite/granite-3.1-2b-instruct/b66dbd045f3e73eb7427.json new file mode 100644 index 0000000000000000000000000000000000000000..e7121b4c906fe4db893266ba29c602f23a655afd --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/granite/ibm-granite/granite-3.1-2b-instruct/b66dbd045f3e73eb7427.json @@ -0,0 +1,72 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "ibm-granite/granite-3.1-2b-instruct", + "_task": "text-generation", + "architectures": [ + "GraniteForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.1, + "attention_multiplier": 0.015625, + "embedding_multiplier": 12.0, + "hidden_act": "silu", + "hidden_size": 2048, + "initializer_range": 0.02, + "intermediate_size": 8192, + "logits_scaling": 8.0, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "granite", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 4, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "ibm-granite/granite-3.1-2b-instruct", + "checkpoint_revision": "bbc2aed595bd38bd770263dc3ab831db9794441d", + "continuous_batching": true, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": true, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 4, + "max_context_length": 4096, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 4096, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": true, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "bfloat16", + "sequence_length": 4096, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "bfloat16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 32, + "num_hidden_layers": 40, + "num_key_value_heads": 8, + "residual_multiplier": 0.22, + "rms_norm_eps": 1e-05, + "rope_scaling": null, + "rope_theta": 5000000.0, + "tie_word_embeddings": true, + "use_cache": true, + "vocab_size": 49155 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/llama/llamafactory/tiny-random-Llama-3/20b7299fc104cc69ce0f.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/llama/llamafactory/tiny-random-Llama-3/20b7299fc104cc69ce0f.json new file mode 100644 index 0000000000000000000000000000000000000000..b99d37fd6f026e6520e51d8391de5ad75d74dcc5 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/llama/llamafactory/tiny-random-Llama-3/20b7299fc104cc69ce0f.json @@ -0,0 +1,76 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "llamafactory/tiny-random-Llama-3", + "_task": "text-generation", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "head_dim": 4, + "hidden_act": "silu", + "hidden_size": 16, + "initializer_range": 0.02, + "intermediate_size": 64, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 1, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "llamafactory/tiny-random-Llama-3", + "checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": true, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 1, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": true, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "bfloat16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "bfloat16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 4, + "num_hidden_layers": 2, + "num_key_value_heads": 4, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 8.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": false, + "use_cache": true, + "vocab_size": 128256 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/llama/llamafactory/tiny-random-Llama-3/502c5523ff7585cd3287.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/llama/llamafactory/tiny-random-Llama-3/502c5523ff7585cd3287.json new file mode 100644 index 0000000000000000000000000000000000000000..182038842fb9c82530fa122c6ef5f6e0522fb392 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/llama/llamafactory/tiny-random-Llama-3/502c5523ff7585cd3287.json @@ -0,0 +1,76 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "llamafactory/tiny-random-Llama-3", + "_task": "text-generation", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "head_dim": 4, + "hidden_act": "silu", + "hidden_size": 16, + "initializer_range": 0.02, + "intermediate_size": 64, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 1, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "llamafactory/tiny-random-Llama-3", + "checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": true, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 1, + "max_context_length": 131072, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 131072, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": true, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "bfloat16", + "sequence_length": 131072, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "bfloat16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 4, + "num_hidden_layers": 2, + "num_key_value_heads": 4, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 8.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": false, + "use_cache": true, + "vocab_size": 128256 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/llama/llamafactory/tiny-random-Llama-3/9c27c3f4a94313192c63.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/llama/llamafactory/tiny-random-Llama-3/9c27c3f4a94313192c63.json new file mode 100644 index 0000000000000000000000000000000000000000..9b7dc60740eb8f34d6a401c5e6f3e38639ebd11b --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/llama/llamafactory/tiny-random-Llama-3/9c27c3f4a94313192c63.json @@ -0,0 +1,76 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "llamafactory/tiny-random-Llama-3", + "_task": "text-generation", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "head_dim": 4, + "hidden_act": "silu", + "hidden_size": 16, + "initializer_range": 0.02, + "intermediate_size": 64, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 1, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "llamafactory/tiny-random-Llama-3", + "checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": true, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 1, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": true, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "float16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "float16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 4, + "num_hidden_layers": 2, + "num_key_value_heads": 4, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 8.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": false, + "use_cache": true, + "vocab_size": 128256 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/llama/llamafactory/tiny-random-Llama-3/f0cd6ae3a57125c25463.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/llama/llamafactory/tiny-random-Llama-3/f0cd6ae3a57125c25463.json new file mode 100644 index 0000000000000000000000000000000000000000..7fc6d6dad78656643bf0bf85bcdff0eafc1b8a4b --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/llama/llamafactory/tiny-random-Llama-3/f0cd6ae3a57125c25463.json @@ -0,0 +1,76 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "llamafactory/tiny-random-Llama-3", + "_task": "text-generation", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "head_dim": 4, + "hidden_act": "silu", + "hidden_size": 16, + "initializer_range": 0.02, + "intermediate_size": 64, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 2, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "llamafactory/tiny-random-Llama-3", + "checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8", + "continuous_batching": true, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": true, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 2, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": true, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "float16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "float16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 4, + "num_hidden_layers": 2, + "num_key_value_heads": 4, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 8.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": false, + "use_cache": true, + "vocab_size": 128256 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/llama/unsloth/Llama-3.2-1B-Instruct/d4cdf18983e9784091ec.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/llama/unsloth/Llama-3.2-1B-Instruct/d4cdf18983e9784091ec.json new file mode 100644 index 0000000000000000000000000000000000000000..5c29f052d524702b8e209cb2328edb9c3f071e2c --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/llama/unsloth/Llama-3.2-1B-Instruct/d4cdf18983e9784091ec.json @@ -0,0 +1,77 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "unsloth/Llama-3.2-1B-Instruct", + "_task": "text-generation", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "head_dim": 64, + "hidden_act": "silu", + "hidden_size": 2048, + "initializer_range": 0.02, + "intermediate_size": 8192, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 4, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "unsloth/Llama-3.2-1B-Instruct", + "checkpoint_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c", + "continuous_batching": true, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": true, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 4, + "max_context_length": 4096, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 4096, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": true, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "float16", + "sequence_length": 4096, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "float16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 32, + "num_hidden_layers": 16, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 32.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": true, + "unsloth_fixed": true, + "use_cache": true, + "vocab_size": 128256 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/mixtral/dacorvo/Mixtral-tiny/34134d6beddf69d6e6c0.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/mixtral/dacorvo/Mixtral-tiny/34134d6beddf69d6e6c0.json new file mode 100644 index 0000000000000000000000000000000000000000..824c7a3c7902d52cedcb662bb940cac654b37b3d --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/mixtral/dacorvo/Mixtral-tiny/34134d6beddf69d6e6c0.json @@ -0,0 +1,72 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "dacorvo/Mixtral-tiny", + "_task": "text-generation", + "architectures": [ + "MixtralForCausalLM" + ], + "attention_dropout": 0.0, + "head_dim": 32, + "hidden_act": "silu", + "hidden_size": 1024, + "initializer_range": 0.02, + "intermediate_size": 3584, + "max_position_embeddings": 1024, + "model_type": "mixtral", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 1, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "dacorvo/Mixtral-tiny", + "checkpoint_revision": "c557ba205ddff6ea911f4719e0d543d6c08356b6", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 1, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": false, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "float16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "float16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 32, + "num_experts_per_tok": 2, + "num_hidden_layers": 2, + "num_key_value_heads": 8, + "num_local_experts": 8, + "output_router_logits": false, + "rms_norm_eps": 1e-05, + "rope_theta": 10000.0, + "router_aux_loss_coef": 0.001, + "router_jitter_noise": 0.0, + "sliding_window": 4096, + "tie_word_embeddings": false, + "use_cache": true, + "vocab_size": 32000 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/mixtral/dacorvo/Mixtral-tiny/4625dfce3a6ec89cfa42.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/mixtral/dacorvo/Mixtral-tiny/4625dfce3a6ec89cfa42.json new file mode 100644 index 0000000000000000000000000000000000000000..a3f8c6cee340a4592d1ff8d3610a87710e44937b --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/mixtral/dacorvo/Mixtral-tiny/4625dfce3a6ec89cfa42.json @@ -0,0 +1,72 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "dacorvo/Mixtral-tiny", + "_task": "text-generation", + "architectures": [ + "MixtralForCausalLM" + ], + "attention_dropout": 0.0, + "head_dim": 32, + "hidden_act": "silu", + "hidden_size": 1024, + "initializer_range": 0.02, + "intermediate_size": 3584, + "max_position_embeddings": 1024, + "model_type": "mixtral", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 1, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "dacorvo/Mixtral-tiny", + "checkpoint_revision": "c557ba205ddff6ea911f4719e0d543d6c08356b6", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 1, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": false, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "bfloat16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "bfloat16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 32, + "num_experts_per_tok": 2, + "num_hidden_layers": 2, + "num_key_value_heads": 8, + "num_local_experts": 8, + "output_router_logits": false, + "rms_norm_eps": 1e-05, + "rope_theta": 10000.0, + "router_aux_loss_coef": 0.001, + "router_jitter_noise": 0.0, + "sliding_window": 4096, + "tie_word_embeddings": false, + "use_cache": true, + "vocab_size": 32000 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/mixtral/dacorvo/Mixtral-tiny/f790967cf96542a801bc.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/mixtral/dacorvo/Mixtral-tiny/f790967cf96542a801bc.json new file mode 100644 index 0000000000000000000000000000000000000000..7ca60b0881e8932521e629ea3474bbdc9c9e2374 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/mixtral/dacorvo/Mixtral-tiny/f790967cf96542a801bc.json @@ -0,0 +1,72 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "dacorvo/Mixtral-tiny", + "_task": "text-generation", + "architectures": [ + "MixtralForCausalLM" + ], + "attention_dropout": 0.0, + "head_dim": 32, + "hidden_act": "silu", + "hidden_size": 1024, + "initializer_range": 0.02, + "intermediate_size": 3584, + "max_position_embeddings": 1024, + "model_type": "mixtral", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 2, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "dacorvo/Mixtral-tiny", + "checkpoint_revision": "c557ba205ddff6ea911f4719e0d543d6c08356b6", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 2, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": false, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "float16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "float16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 32, + "num_experts_per_tok": 2, + "num_hidden_layers": 2, + "num_key_value_heads": 8, + "num_local_experts": 8, + "output_router_logits": false, + "rms_norm_eps": 1e-05, + "rope_theta": 10000.0, + "router_aux_loss_coef": 0.001, + "router_jitter_noise": 0.0, + "sliding_window": 4096, + "tie_word_embeddings": false, + "use_cache": true, + "vocab_size": 32000 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/phi3/microsoft/Phi-3-mini-4k-instruct/c3b4f930a6c51ad40028.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/phi3/microsoft/Phi-3-mini-4k-instruct/c3b4f930a6c51ad40028.json new file mode 100644 index 0000000000000000000000000000000000000000..0b68023b3256667bbac3b5f67de4459529326bf9 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/phi3/microsoft/Phi-3-mini-4k-instruct/c3b4f930a6c51ad40028.json @@ -0,0 +1,76 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "microsoft/Phi-3-mini-4k-instruct", + "_task": "text-generation", + "architectures": [ + "Phi3ForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "auto_map": { + "AutoConfig": "configuration_phi3.Phi3Config", + "AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM" + }, + "embd_pdrop": 0.0, + "hidden_act": "silu", + "hidden_size": 3072, + "initializer_range": 0.02, + "intermediate_size": 8192, + "max_position_embeddings": 4096, + "model_type": "phi3", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 4, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "microsoft/Phi-3-mini-4k-instruct", + "checkpoint_revision": "0a67737cc96d2554230f90338b163bc6380a2a85", + "continuous_batching": true, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": true, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 4, + "max_context_length": 4096, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 4096, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": true, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "bfloat16", + "sequence_length": 4096, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "bfloat16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 32, + "num_hidden_layers": 32, + "num_key_value_heads": 32, + "original_max_position_embeddings": 4096, + "partial_rotary_factor": 1.0, + "resid_pdrop": 0.0, + "rms_norm_eps": 1e-05, + "rope_scaling": null, + "rope_theta": 10000.0, + "sliding_window": 2047, + "tie_word_embeddings": false, + "use_cache": true, + "vocab_size": 32064 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/phi3/yujiepan/phi-4-tiny-random/63d7815d3de40f55a3ae.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/phi3/yujiepan/phi-4-tiny-random/63d7815d3de40f55a3ae.json new file mode 100644 index 0000000000000000000000000000000000000000..961a8ebebe5bcc656df88a7d8b4d6b22264bd968 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/phi3/yujiepan/phi-4-tiny-random/63d7815d3de40f55a3ae.json @@ -0,0 +1,73 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "yujiepan/phi-4-tiny-random", + "_task": "text-generation", + "architectures": [ + "Phi3ForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "auto_map": {}, + "embd_pdrop": 0.0, + "hidden_act": "silu", + "hidden_size": 16, + "initializer_range": 0.02, + "intermediate_size": 32, + "max_position_embeddings": 16384, + "model_type": "phi3", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 2, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "yujiepan/phi-4-tiny-random", + "checkpoint_revision": "18a9a1168dc97ac6d128f811925670c275610f5a", + "continuous_batching": true, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": true, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 2, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": true, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "float16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "float16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 2, + "num_hidden_layers": 2, + "num_key_value_heads": 1, + "original_max_position_embeddings": 16384, + "partial_rotary_factor": 1.0, + "resid_pdrop": 0.0, + "rms_norm_eps": 1e-05, + "rope_scaling": null, + "rope_theta": 250000, + "sliding_window": null, + "tie_word_embeddings": false, + "use_cache": true, + "vocab_size": 100352 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/phi3/yujiepan/phi-4-tiny-random/892d9f2862662e5407c6.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/phi3/yujiepan/phi-4-tiny-random/892d9f2862662e5407c6.json new file mode 100644 index 0000000000000000000000000000000000000000..743f1c049614d46665c8b689664ab7a761d7d8fc --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/phi3/yujiepan/phi-4-tiny-random/892d9f2862662e5407c6.json @@ -0,0 +1,73 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "yujiepan/phi-4-tiny-random", + "_task": "text-generation", + "architectures": [ + "Phi3ForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "auto_map": {}, + "embd_pdrop": 0.0, + "hidden_act": "silu", + "hidden_size": 16, + "initializer_range": 0.02, + "intermediate_size": 32, + "max_position_embeddings": 16384, + "model_type": "phi3", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 1, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "yujiepan/phi-4-tiny-random", + "checkpoint_revision": "18a9a1168dc97ac6d128f811925670c275610f5a", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": true, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 1, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": true, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "float16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "float16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 2, + "num_hidden_layers": 2, + "num_key_value_heads": 1, + "original_max_position_embeddings": 16384, + "partial_rotary_factor": 1.0, + "resid_pdrop": 0.0, + "rms_norm_eps": 1e-05, + "rope_scaling": null, + "rope_theta": 250000, + "sliding_window": null, + "tie_word_embeddings": false, + "use_cache": true, + "vocab_size": 100352 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/phi3/yujiepan/phi-4-tiny-random/e97f2d9bfc450d08ef3b.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/phi3/yujiepan/phi-4-tiny-random/e97f2d9bfc450d08ef3b.json new file mode 100644 index 0000000000000000000000000000000000000000..5fbe5d6bbb64d936c0fe916341bf36605a0e8e56 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/phi3/yujiepan/phi-4-tiny-random/e97f2d9bfc450d08ef3b.json @@ -0,0 +1,73 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "yujiepan/phi-4-tiny-random", + "_task": "text-generation", + "architectures": [ + "Phi3ForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "auto_map": {}, + "embd_pdrop": 0.0, + "hidden_act": "silu", + "hidden_size": 16, + "initializer_range": 0.02, + "intermediate_size": 32, + "max_position_embeddings": 16384, + "model_type": "phi3", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 1, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "yujiepan/phi-4-tiny-random", + "checkpoint_revision": "18a9a1168dc97ac6d128f811925670c275610f5a", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": true, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 1, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": true, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "bfloat16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "bfloat16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 2, + "num_hidden_layers": 2, + "num_key_value_heads": 1, + "original_max_position_embeddings": 16384, + "partial_rotary_factor": 1.0, + "resid_pdrop": 0.0, + "rms_norm_eps": 1e-05, + "rope_scaling": null, + "rope_theta": 250000, + "sliding_window": null, + "tie_word_embeddings": false, + "use_cache": true, + "vocab_size": 100352 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen2/Qwen/Qwen2.5-0.5B/00b909b9addfb3e82c75.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen2/Qwen/Qwen2.5-0.5B/00b909b9addfb3e82c75.json new file mode 100644 index 0000000000000000000000000000000000000000..29a48eac91b0b227155f6c38e27c364a85cc2d76 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen2/Qwen/Qwen2.5-0.5B/00b909b9addfb3e82c75.json @@ -0,0 +1,96 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "Qwen/Qwen2.5-0.5B", + "_task": "text-generation", + "architectures": [ + "Qwen2ForCausalLM" + ], + "attention_dropout": 0.0, + "hidden_act": "silu", + "hidden_size": 896, + "initializer_range": 0.02, + "intermediate_size": 4864, + "layer_types": [ + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention" + ], + "max_position_embeddings": 32768, + "max_window_layers": 24, + "model_type": "qwen2", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 4, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "Qwen/Qwen2.5-0.5B", + "checkpoint_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987", + "continuous_batching": true, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 4, + "max_context_length": 4096, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 4096, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": false, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "float16", + "sequence_length": 4096, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "float16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 14, + "num_hidden_layers": 24, + "num_key_value_heads": 2, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 1000000.0, + "sliding_window": null, + "tie_word_embeddings": true, + "use_cache": true, + "use_mrope": false, + "use_sliding_window": false, + "vocab_size": 151936 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen2/Qwen/Qwen2.5-0.5B/8a714805c0774dccda3d.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen2/Qwen/Qwen2.5-0.5B/8a714805c0774dccda3d.json new file mode 100644 index 0000000000000000000000000000000000000000..5c3e387f990a3e8e7dc3def8b548bf22b3a9a39b --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen2/Qwen/Qwen2.5-0.5B/8a714805c0774dccda3d.json @@ -0,0 +1,96 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "Qwen/Qwen2.5-0.5B", + "_task": "text-generation", + "architectures": [ + "Qwen2ForCausalLM" + ], + "attention_dropout": 0.0, + "hidden_act": "silu", + "hidden_size": 896, + "initializer_range": 0.02, + "intermediate_size": 4864, + "layer_types": [ + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention" + ], + "max_position_embeddings": 32768, + "max_window_layers": 24, + "model_type": "qwen2", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 1, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "Qwen/Qwen2.5-0.5B", + "checkpoint_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 1, + "max_context_length": 32768, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 32768, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": true, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "bfloat16", + "sequence_length": 32768, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "bfloat16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 14, + "num_hidden_layers": 24, + "num_key_value_heads": 2, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 1000000.0, + "sliding_window": null, + "tie_word_embeddings": true, + "use_cache": true, + "use_mrope": false, + "use_sliding_window": false, + "vocab_size": 151936 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen2/yujiepan/qwen2.5-128k-tiny-random/4be7e684c15f704cb67d.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen2/yujiepan/qwen2.5-128k-tiny-random/4be7e684c15f704cb67d.json new file mode 100644 index 0000000000000000000000000000000000000000..9b4e06bfad0de66d9cdf9613baca71f012b08e66 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen2/yujiepan/qwen2.5-128k-tiny-random/4be7e684c15f704cb67d.json @@ -0,0 +1,78 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "yujiepan/qwen2.5-128k-tiny-random", + "_task": "text-generation", + "architectures": [ + "Qwen2ForCausalLM" + ], + "attention_dropout": 0.0, + "hidden_act": "silu", + "hidden_size": 8, + "initializer_range": 0.02, + "intermediate_size": 16, + "layer_types": [ + "full_attention", + "full_attention" + ], + "max_position_embeddings": 32768, + "max_window_layers": 1, + "model_type": "qwen2", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 2, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "yujiepan/qwen2.5-128k-tiny-random", + "checkpoint_revision": "c8296d4ca3f87782876d2382fbb6481d1beb8ef0", + "continuous_batching": true, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 2, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": false, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "float16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "float16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 4, + "num_hidden_layers": 2, + "num_key_value_heads": 2, + "rms_norm_eps": 1e-06, + "rope_scaling": { + "factor": 4.0, + "original_max_position_embeddings": 32768, + "rope_type": "yarn", + "type": "yarn" + }, + "rope_theta": 1000000.0, + "sliding_window": null, + "tie_word_embeddings": false, + "use_cache": true, + "use_sliding_window": false, + "vocab_size": 152064 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen2/yujiepan/qwen2.5-128k-tiny-random/6cead920a0dbb3daefb9.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen2/yujiepan/qwen2.5-128k-tiny-random/6cead920a0dbb3daefb9.json new file mode 100644 index 0000000000000000000000000000000000000000..73a3837cd498485e0e431571d66338f428215170 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen2/yujiepan/qwen2.5-128k-tiny-random/6cead920a0dbb3daefb9.json @@ -0,0 +1,78 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "yujiepan/qwen2.5-128k-tiny-random", + "_task": "text-generation", + "architectures": [ + "Qwen2ForCausalLM" + ], + "attention_dropout": 0.0, + "hidden_act": "silu", + "hidden_size": 8, + "initializer_range": 0.02, + "intermediate_size": 16, + "layer_types": [ + "full_attention", + "full_attention" + ], + "max_position_embeddings": 32768, + "max_window_layers": 1, + "model_type": "qwen2", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 1, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "yujiepan/qwen2.5-128k-tiny-random", + "checkpoint_revision": "c8296d4ca3f87782876d2382fbb6481d1beb8ef0", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 1, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": true, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "float16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "float16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 4, + "num_hidden_layers": 2, + "num_key_value_heads": 2, + "rms_norm_eps": 1e-06, + "rope_scaling": { + "factor": 4.0, + "original_max_position_embeddings": 32768, + "rope_type": "yarn", + "type": "yarn" + }, + "rope_theta": 1000000.0, + "sliding_window": null, + "tie_word_embeddings": false, + "use_cache": true, + "use_sliding_window": false, + "vocab_size": 152064 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen2/yujiepan/qwen2.5-128k-tiny-random/cd761b198c1bec21bd55.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen2/yujiepan/qwen2.5-128k-tiny-random/cd761b198c1bec21bd55.json new file mode 100644 index 0000000000000000000000000000000000000000..a48e6102aa99758b5efa2fa53381a6c5f6ed3e2b --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen2/yujiepan/qwen2.5-128k-tiny-random/cd761b198c1bec21bd55.json @@ -0,0 +1,78 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "yujiepan/qwen2.5-128k-tiny-random", + "_task": "text-generation", + "architectures": [ + "Qwen2ForCausalLM" + ], + "attention_dropout": 0.0, + "hidden_act": "silu", + "hidden_size": 8, + "initializer_range": 0.02, + "intermediate_size": 16, + "layer_types": [ + "full_attention", + "full_attention" + ], + "max_position_embeddings": 32768, + "max_window_layers": 1, + "model_type": "qwen2", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 1, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "yujiepan/qwen2.5-128k-tiny-random", + "checkpoint_revision": "c8296d4ca3f87782876d2382fbb6481d1beb8ef0", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 1, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": true, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "bfloat16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "bfloat16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 4, + "num_hidden_layers": 2, + "num_key_value_heads": 2, + "rms_norm_eps": 1e-06, + "rope_scaling": { + "factor": 4.0, + "original_max_position_embeddings": 32768, + "rope_type": "yarn", + "type": "yarn" + }, + "rope_theta": 1000000.0, + "sliding_window": null, + "tie_word_embeddings": false, + "use_cache": true, + "use_sliding_window": false, + "vocab_size": 152064 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen3/Qwen/Qwen3-0.6B/2ed7a8812dbe9c7ab058.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen3/Qwen/Qwen3-0.6B/2ed7a8812dbe9c7ab058.json new file mode 100644 index 0000000000000000000000000000000000000000..b4644255a9dac32dd8eb00e9ac0cdf5295f62dc6 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen3/Qwen/Qwen3-0.6B/2ed7a8812dbe9c7ab058.json @@ -0,0 +1,101 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "Qwen/Qwen3-0.6B", + "_task": "text-generation", + "architectures": [ + "Qwen3ForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 1024, + "initializer_range": 0.02, + "intermediate_size": 3072, + "layer_types": [ + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention" + ], + "max_position_embeddings": 40960, + "max_window_layers": 28, + "model_type": "qwen3", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 4, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "Qwen/Qwen3-0.6B", + "checkpoint_revision": "c1899de289a04d12100db370d81485cdf75e47ca", + "continuous_batching": true, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": true, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 4, + "max_context_length": 4096, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 4096, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": false, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "bfloat16", + "sequence_length": 4096, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "bfloat16", + "tp_degree": 2, + "vocab_parallel": false + }, + "num_attention_heads": 16, + "num_hidden_layers": 28, + "num_key_value_heads": 8, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 1000000, + "sliding_window": null, + "tie_word_embeddings": true, + "use_cache": true, + "use_sliding_window": false, + "vocab_size": 151936 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/8fff447965fd3c157c0f.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/8fff447965fd3c157c0f.json new file mode 100644 index 0000000000000000000000000000000000000000..48db4015486e3d22d008ca6ae1347c80498f2e90 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/8fff447965fd3c157c0f.json @@ -0,0 +1,79 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "optimum-internal-testing/tiny-random-qwen3_moe", + "_task": "text-generation", + "architectures": [ + "Qwen3MoeForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "decoder_sparse_step": 2, + "head_dim": 32, + "hidden_act": "silu", + "hidden_size": 64, + "initializer_range": 0.02, + "intermediate_size": 128, + "max_position_embeddings": 40960, + "max_window_layers": 1, + "mlp_only_layers": [], + "model_type": "qwen3_moe", + "moe_intermediate_size": 128, + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 2, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "optimum-internal-testing/tiny-random-qwen3_moe", + "checkpoint_revision": "e0230be2839556b44b7400a233c73c74b4abb7af", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 2, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": false, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "float16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "float16", + "tp_degree": 2, + "vocab_parallel": false + }, + "norm_topk_prob": true, + "num_attention_heads": 2, + "num_experts": 8, + "num_experts_per_tok": 2, + "num_hidden_layers": 2, + "num_key_value_heads": 1, + "output_router_logits": false, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 1000000.0, + "router_aux_loss_coef": 0.001, + "sliding_window": null, + "tie_word_embeddings": true, + "use_cache": true, + "use_sliding_window": false, + "vocab_size": 151936 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/90f9e02faff8566070cb.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/90f9e02faff8566070cb.json new file mode 100644 index 0000000000000000000000000000000000000000..439d714c09bc2a2132aa6f1c4139d4de90cc210e --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/90f9e02faff8566070cb.json @@ -0,0 +1,79 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "optimum-internal-testing/tiny-random-qwen3_moe", + "_task": "text-generation", + "architectures": [ + "Qwen3MoeForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "decoder_sparse_step": 2, + "head_dim": 32, + "hidden_act": "silu", + "hidden_size": 64, + "initializer_range": 0.02, + "intermediate_size": 128, + "max_position_embeddings": 40960, + "max_window_layers": 1, + "mlp_only_layers": [], + "model_type": "qwen3_moe", + "moe_intermediate_size": 128, + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 1, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "optimum-internal-testing/tiny-random-qwen3_moe", + "checkpoint_revision": "e0230be2839556b44b7400a233c73c74b4abb7af", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 1, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": false, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "float16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "float16", + "tp_degree": 2, + "vocab_parallel": false + }, + "norm_topk_prob": true, + "num_attention_heads": 2, + "num_experts": 8, + "num_experts_per_tok": 2, + "num_hidden_layers": 2, + "num_key_value_heads": 1, + "output_router_logits": false, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 1000000.0, + "router_aux_loss_coef": 0.001, + "sliding_window": null, + "tie_word_embeddings": true, + "use_cache": true, + "use_sliding_window": false, + "vocab_size": 151936 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/f6c02365716b13dea692.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/f6c02365716b13dea692.json new file mode 100644 index 0000000000000000000000000000000000000000..fe71b4c5cca937e3b71a2e2935dd307755730cec --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/f6c02365716b13dea692.json @@ -0,0 +1,79 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "optimum-internal-testing/tiny-random-qwen3_moe", + "_task": "text-generation", + "architectures": [ + "Qwen3MoeForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "decoder_sparse_step": 2, + "head_dim": 32, + "hidden_act": "silu", + "hidden_size": 64, + "initializer_range": 0.02, + "intermediate_size": 128, + "max_position_embeddings": 40960, + "max_window_layers": 1, + "mlp_only_layers": [], + "model_type": "qwen3_moe", + "moe_intermediate_size": 128, + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 1, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "optimum-internal-testing/tiny-random-qwen3_moe", + "checkpoint_revision": "e0230be2839556b44b7400a233c73c74b4abb7af", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 1, + "max_context_length": 100, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 100, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": false, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "bfloat16", + "sequence_length": 100, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "bfloat16", + "tp_degree": 2, + "vocab_parallel": false + }, + "norm_topk_prob": true, + "num_attention_heads": 2, + "num_experts": 8, + "num_experts_per_tok": 2, + "num_hidden_layers": 2, + "num_key_value_heads": 1, + "output_router_logits": false, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 1000000.0, + "router_aux_loss_coef": 0.001, + "sliding_window": null, + "tie_word_embeddings": true, + "use_cache": true, + "use_sliding_window": false, + "vocab_size": 151936 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/smollm3/HuggingFaceTB/SmolLM3-3B/60de13af0adf5a679b2c.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/smollm3/HuggingFaceTB/SmolLM3-3B/60de13af0adf5a679b2c.json new file mode 100644 index 0000000000000000000000000000000000000000..bb9bd7391e9304e814cce740a5cb70f5df13d560 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev2/smollm3/HuggingFaceTB/SmolLM3-3B/60de13af0adf5a679b2c.json @@ -0,0 +1,148 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "HuggingFaceTB/SmolLM3-3B", + "_task": "text-generation", + "architectures": [ + "SmolLM3ForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "hidden_act": "silu", + "hidden_size": 2048, + "initializer_range": 0.02, + "intermediate_size": 11008, + "layer_types": [ + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention" + ], + "max_position_embeddings": 65536, + "max_window_layers": 28, + "mlp_bias": false, + "model_type": "smollm3", + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 4, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "HuggingFaceTB/SmolLM3-3B", + "checkpoint_revision": "1c00fc78bd9cf90108046bc433cb34992480f1c1", + "continuous_batching": true, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": true, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 2, + "logical_nc_config": 1, + "max_batch_size": 4, + "max_context_length": 4096, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 4096, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": true, + "optimum_neuron_version": "0.3.1.dev2", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "bfloat16", + "sequence_length": 4096, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "bfloat16", + "tp_degree": 2, + "vocab_parallel": false + }, + "no_rope_layer_interval": 4, + "no_rope_layers": [ + 1, + 1, + 1, + 0, + 1, + 1, + 1, + 0, + 1, + 1, + 1, + 0, + 1, + 1, + 1, + 0, + 1, + 1, + 1, + 0, + 1, + 1, + 1, + 0, + 1, + 1, + 1, + 0, + 1, + 1, + 1, + 0, + 1, + 1, + 1, + 0 + ], + "num_attention_heads": 16, + "num_hidden_layers": 36, + "num_key_value_heads": 4, + "pretraining_tp": 2, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 5000000.0, + "sliding_window": null, + "use_cache": false, + "use_sliding_window": false, + "vocab_size": 128256 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev3/qwen3_moe/Qwen/Qwen3-30B-A3B-Instruct-2507/fc72a21b6de27e9bcefe.json b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev3/qwen3_moe/Qwen/Qwen3-30B-A3B-Instruct-2507/fc72a21b6de27e9bcefe.json new file mode 100644 index 0000000000000000000000000000000000000000..655e2f4718c56c730b9c7018ea7085235388274d --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev3/qwen3_moe/Qwen/Qwen3-30B-A3B-Instruct-2507/fc72a21b6de27e9bcefe.json @@ -0,0 +1,79 @@ +{ + "_entry_class": "SingleModelCacheEntry", + "_model_id": "Qwen/Qwen3-30B-A3B-Instruct-2507", + "_task": "text-generation", + "architectures": [ + "Qwen3MoeForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "decoder_sparse_step": 1, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 2048, + "initializer_range": 0.02, + "intermediate_size": 6144, + "max_position_embeddings": 262144, + "max_window_layers": 48, + "mlp_only_layers": [], + "model_type": "qwen3_moe", + "moe_intermediate_size": 768, + "neuron": { + "_serialized_key": "NxDNeuronConfig", + "async_mode": false, + "attn_kernel_enabled": false, + "batch_size": 8, + "capacity_factor": null, + "cc_pipeline_tiling_factor": 2, + "checkpoint_id": "Qwen/Qwen3-30B-A3B-Instruct-2507", + "checkpoint_revision": "61082d4deaa4785f64943b443cbc2b5de7524fad", + "continuous_batching": false, + "enable_bucketing": false, + "ep_degree": 1, + "flash_decoding_enabled": false, + "fused_qkv": false, + "glu_mlp": true, + "is_chunked_prefill": false, + "local_ranks_size": 8, + "logical_nc_config": 1, + "max_batch_size": 8, + "max_context_length": 4096, + "max_topk": 256, + "mlp_kernel_enabled": false, + "mlp_kernel_fuse_residual_add": false, + "n_active_tokens": 4096, + "neuronxcc_version": "2.19.8089.0+8ab9f450", + "num_cores_per_group": 1, + "on_device_sampling": false, + "optimum_neuron_version": "0.3.1.dev3", + "output_logits": false, + "padding_side": "right", + "pp_degree": 1, + "qkv_kernel_enabled": false, + "rpl_reduce_dtype": "bfloat16", + "sequence_length": 4096, + "sequence_parallel_enabled": false, + "speculation_length": 0, + "start_rank_id": 0, + "target": null, + "torch_dtype": "bfloat16", + "tp_degree": 8, + "vocab_parallel": false + }, + "norm_topk_prob": true, + "num_attention_heads": 32, + "num_experts": 128, + "num_experts_per_tok": 8, + "num_hidden_layers": 48, + "num_key_value_heads": 4, + "output_router_logits": false, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 10000000, + "router_aux_loss_coef": 0.001, + "sliding_window": null, + "tie_word_embeddings": false, + "use_cache": true, + "use_sliding_window": false, + "vocab_size": 151936 +} \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_04376e1307045cf435ae+431f5505/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_04376e1307045cf435ae+431f5505/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..9932c70d27d759d781be88485b74b17b49b4c04f --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_04376e1307045cf435ae+431f5505/compile_flags.json @@ -0,0 +1 @@ +"--model-type=transformer -O1 --lnc=1 --internal-hlo2tensorizer-options=--experimental-unsafe-fp8e4m3fn-as-fp8e4m3 --logfile=/tmp/nxd_model/layout_opt/log-neuron-cc.txt" \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_04376e1307045cf435ae+431f5505/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_04376e1307045cf435ae+431f5505/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_04376e1307045cf435ae+431f5505/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_04376e1307045cf435ae+431f5505/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..ff7d986b3271d37835d84abc55a8a39dfcf714eb --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_04376e1307045cf435ae+431f5505/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1a6a5e05e7901f7ec6ea776377316585b1bf917c33884f8a2a2d3851e28128c6 +size 7993 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_04376e1307045cf435ae+431f5505/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_04376e1307045cf435ae+431f5505/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..e1a9110e65f3dfe16914ed5de35c393ff98056b2 Binary files /dev/null and b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_04376e1307045cf435ae+431f5505/model.neff differ diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0a47deab436eaf26c99a+cd3419b6/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0a47deab436eaf26c99a+cd3419b6/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..93354bd42abb66b9b6e56b127441174baa403534 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0a47deab436eaf26c99a+cd3419b6/compile_flags.json @@ -0,0 +1 @@ +["--target=trn1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"] \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0a47deab436eaf26c99a+cd3419b6/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0a47deab436eaf26c99a+cd3419b6/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0a47deab436eaf26c99a+cd3419b6/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0a47deab436eaf26c99a+cd3419b6/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..6a9334b0b848beed885ab182d0e274d3e3cb9431 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0a47deab436eaf26c99a+cd3419b6/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a31c4cdb1fbc97738a6e37dec33d6f44e6b1284cef02f7149efa6cfbe91b480c +size 63836 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0a47deab436eaf26c99a+cd3419b6/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0a47deab436eaf26c99a+cd3419b6/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..87a31198f24982c87763248012545bda30bcd43f --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0a47deab436eaf26c99a+cd3419b6/model.neff @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8842daaea9ad716443a64507d5e6eb80bd5036701c69d3cc22d3c8b28cfe3129 +size 297984 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0a47deab436eaf26c99a+cd3419b6/wrapped_neff.hlo b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0a47deab436eaf26c99a+cd3419b6/wrapped_neff.hlo new file mode 100644 index 0000000000000000000000000000000000000000..e38e58c3c706b254edb6a0ae9c697677338cd49a --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0a47deab436eaf26c99a+cd3419b6/wrapped_neff.hlo @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:01904b235c4fe11e10878eb7201accabe15e04a5a33cb5e989bab9b2b9fd2617 +size 307570 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0b67a734fc55d67768bd+253d6470/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0b67a734fc55d67768bd+253d6470/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..cb53ce53583cfe21339b4bed39f5fa1a679cddd1 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0b67a734fc55d67768bd+253d6470/compile_flags.json @@ -0,0 +1 @@ +["--target=trn1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"] \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0b67a734fc55d67768bd+253d6470/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0b67a734fc55d67768bd+253d6470/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0b67a734fc55d67768bd+253d6470/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0b67a734fc55d67768bd+253d6470/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..80d9136200b58e6269150c051fd495b5aaeee849 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0b67a734fc55d67768bd+253d6470/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a0f136968c19fe7f46131dfedad1594b894b6655dc565cdc5558de5cacfb3f05 +size 79740 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0b67a734fc55d67768bd+253d6470/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0b67a734fc55d67768bd+253d6470/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..85ae2f5300dd2610c8f7fe594924b7bea675a3a0 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_0b67a734fc55d67768bd+253d6470/model.neff @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:82cf7b5f3ae04b99b054cf93a92f9347d6a7d0e73783fedf2611a11b2d3fbe2a +size 297984 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_15bb8d9c0246df177b49+431f5505/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_15bb8d9c0246df177b49+431f5505/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..9932c70d27d759d781be88485b74b17b49b4c04f --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_15bb8d9c0246df177b49+431f5505/compile_flags.json @@ -0,0 +1 @@ +"--model-type=transformer -O1 --lnc=1 --internal-hlo2tensorizer-options=--experimental-unsafe-fp8e4m3fn-as-fp8e4m3 --logfile=/tmp/nxd_model/layout_opt/log-neuron-cc.txt" \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_15bb8d9c0246df177b49+431f5505/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_15bb8d9c0246df177b49+431f5505/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_15bb8d9c0246df177b49+431f5505/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_15bb8d9c0246df177b49+431f5505/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..b52d8cc266e17bc530bcad08196823405bf20df5 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_15bb8d9c0246df177b49+431f5505/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:41d96c35c2ccaa4ae59a468b8fd21975e79107d8066114014997dc59ecebd487 +size 9389 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_15bb8d9c0246df177b49+431f5505/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_15bb8d9c0246df177b49+431f5505/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..3b56c0796bac36c24ff05da60f9cb8950a473d08 Binary files /dev/null and b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_15bb8d9c0246df177b49+431f5505/model.neff differ diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_184fcc045921c74845be+cd3419b6/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_184fcc045921c74845be+cd3419b6/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..93354bd42abb66b9b6e56b127441174baa403534 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_184fcc045921c74845be+cd3419b6/compile_flags.json @@ -0,0 +1 @@ +["--target=trn1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"] \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_184fcc045921c74845be+cd3419b6/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_184fcc045921c74845be+cd3419b6/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_184fcc045921c74845be+cd3419b6/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_184fcc045921c74845be+cd3419b6/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..85fbf4a25cf0195c367f8d929722f97674d63676 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_184fcc045921c74845be+cd3419b6/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0b06a008527980f356db3d9da2f8e07a0f666e78a972b1f132cf73888b442c2e +size 74871 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_184fcc045921c74845be+cd3419b6/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_184fcc045921c74845be+cd3419b6/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..b83acec29a32a4d122cbac0dd7e5fb10ae9299dc --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_184fcc045921c74845be+cd3419b6/model.neff @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ecc6293b1f4e819e5c75dc187918683fae55afcde0a88804cc55d9368db84364 +size 287744 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_184fcc045921c74845be+cd3419b6/wrapped_neff.hlo b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_184fcc045921c74845be+cd3419b6/wrapped_neff.hlo new file mode 100644 index 0000000000000000000000000000000000000000..17962bfa590136ed19690d8650f5998f72b00b1c --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_184fcc045921c74845be+cd3419b6/wrapped_neff.hlo @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:03e64756b32beefb88ac3aa57215d382b1b191f8f8299ea04d176bfd32d6e906 +size 297333 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1940237707648304797+e30acd3a/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1940237707648304797+e30acd3a/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..32157f229be17480afc9e02a7ed48ca0cbc8f22a --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1940237707648304797+e30acd3a/compile_flags.json @@ -0,0 +1 @@ +["--target=trn1"] \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1940237707648304797+e30acd3a/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1940237707648304797+e30acd3a/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1940237707648304797+e30acd3a/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1940237707648304797+e30acd3a/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..6ec09fc8d2297f6e534d45ab48b9115a3ec449cd --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1940237707648304797+e30acd3a/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:54c474fb284f0e1cdb6dc86b55c91b03eaf8f0421d872dddec4becec301c1956 +size 970 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1940237707648304797+e30acd3a/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1940237707648304797+e30acd3a/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..d7f3de257aa981beebc0f37b320a039c900b84f9 Binary files /dev/null and b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1940237707648304797+e30acd3a/model.neff differ diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1c6882381e4240eeeb0c+431f5505/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1c6882381e4240eeeb0c+431f5505/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..9932c70d27d759d781be88485b74b17b49b4c04f --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1c6882381e4240eeeb0c+431f5505/compile_flags.json @@ -0,0 +1 @@ +"--model-type=transformer -O1 --lnc=1 --internal-hlo2tensorizer-options=--experimental-unsafe-fp8e4m3fn-as-fp8e4m3 --logfile=/tmp/nxd_model/layout_opt/log-neuron-cc.txt" \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1c6882381e4240eeeb0c+431f5505/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1c6882381e4240eeeb0c+431f5505/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1c6882381e4240eeeb0c+431f5505/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1c6882381e4240eeeb0c+431f5505/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..f1e6668ec2317f34708baa0e99995b8a6a6a57ee --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1c6882381e4240eeeb0c+431f5505/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:78e3e7c5cdd402819f82ab95b0d26d53bc9ee895fc8dd5621fdae8ea91257f1d +size 10039 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1c6882381e4240eeeb0c+431f5505/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1c6882381e4240eeeb0c+431f5505/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..173ac6d76aa2ccf688bd3526b33c0dfb2cfb2eff Binary files /dev/null and b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1c6882381e4240eeeb0c+431f5505/model.neff differ diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1d0c9d2357e7888e46cd+253d6470/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1d0c9d2357e7888e46cd+253d6470/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..cb53ce53583cfe21339b4bed39f5fa1a679cddd1 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1d0c9d2357e7888e46cd+253d6470/compile_flags.json @@ -0,0 +1 @@ +["--target=trn1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"] \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1d0c9d2357e7888e46cd+253d6470/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1d0c9d2357e7888e46cd+253d6470/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1d0c9d2357e7888e46cd+253d6470/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1d0c9d2357e7888e46cd+253d6470/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..c82183d4432394117c59cdf1ca01cfa4bfcca234 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1d0c9d2357e7888e46cd+253d6470/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d4b5a6b6fb9269ad0e90c9d158fdfe0d144a65523095d36430cdd75a94f310aa +size 80577 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1d0c9d2357e7888e46cd+253d6470/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1d0c9d2357e7888e46cd+253d6470/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..f4c41bb44134dba18e822d60a878edd83363b7e3 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_1d0c9d2357e7888e46cd+253d6470/model.neff @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fbe503446ae4be16ba5818c37cddd064d2c658ceb24a0aa88e61311317846a4a +size 277504 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_20ce63f489314924a057+253d6470/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_20ce63f489314924a057+253d6470/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..cb53ce53583cfe21339b4bed39f5fa1a679cddd1 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_20ce63f489314924a057+253d6470/compile_flags.json @@ -0,0 +1 @@ +["--target=trn1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"] \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_20ce63f489314924a057+253d6470/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_20ce63f489314924a057+253d6470/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_20ce63f489314924a057+253d6470/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_20ce63f489314924a057+253d6470/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..f73f6201536fb8b2294cf168aa9e8d4d2bb7e37c --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_20ce63f489314924a057+253d6470/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:adf143da6ef102c634a18ed6bc5782ab446f0272f4b281d2280b27019b18b923 +size 63331 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_20ce63f489314924a057+253d6470/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_20ce63f489314924a057+253d6470/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..15bb877b84420ac1714c5b7825ad0b880e1be981 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_20ce63f489314924a057+253d6470/model.neff @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cce35b0fbb4a24282647ff1dd7183bb4cc0dd6857c6b0d7149628849d24258a +size 287744 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_2d2de0754e92d084bfdd+431f5505/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_2d2de0754e92d084bfdd+431f5505/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..9932c70d27d759d781be88485b74b17b49b4c04f --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_2d2de0754e92d084bfdd+431f5505/compile_flags.json @@ -0,0 +1 @@ +"--model-type=transformer -O1 --lnc=1 --internal-hlo2tensorizer-options=--experimental-unsafe-fp8e4m3fn-as-fp8e4m3 --logfile=/tmp/nxd_model/layout_opt/log-neuron-cc.txt" \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_2d2de0754e92d084bfdd+431f5505/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_2d2de0754e92d084bfdd+431f5505/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_2d2de0754e92d084bfdd+431f5505/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_2d2de0754e92d084bfdd+431f5505/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..221b265a1790316640ea59d67d770f52ff0c3d9c --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_2d2de0754e92d084bfdd+431f5505/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3c771665be0ee7502116b8ce0f5a91ff36d81b629eea10fb3ca2b7ab256b87fb +size 10039 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_2d2de0754e92d084bfdd+431f5505/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_2d2de0754e92d084bfdd+431f5505/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..47391292d14ecb7e312a19b19a92c86a8e034fba Binary files /dev/null and b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_2d2de0754e92d084bfdd+431f5505/model.neff differ diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_40f6b4304be01d72d956+253d6470/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_40f6b4304be01d72d956+253d6470/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..cb53ce53583cfe21339b4bed39f5fa1a679cddd1 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_40f6b4304be01d72d956+253d6470/compile_flags.json @@ -0,0 +1 @@ +["--target=trn1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"] \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_40f6b4304be01d72d956+253d6470/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_40f6b4304be01d72d956+253d6470/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_40f6b4304be01d72d956+253d6470/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_40f6b4304be01d72d956+253d6470/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..368b0ae888ec07c27b637793414d7d4e58eaf72c --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_40f6b4304be01d72d956+253d6470/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0313ab33efa7bc2afc16e69a56885768329bd7f5234cf53905ccc9a505bbdc9c +size 300195 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_40f6b4304be01d72d956+253d6470/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_40f6b4304be01d72d956+253d6470/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..d11cc9e37e97e7984b540308540694a761aa7ac6 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_40f6b4304be01d72d956+253d6470/model.neff @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3b189ee48bca944eb76c443beeb3e74126739c3cf8d909f76b4ccb01e12fec76 +size 461824 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_64f5633fb25cfe2548d6+253d6470/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_64f5633fb25cfe2548d6+253d6470/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..cb53ce53583cfe21339b4bed39f5fa1a679cddd1 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_64f5633fb25cfe2548d6+253d6470/compile_flags.json @@ -0,0 +1 @@ +["--target=trn1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"] \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_64f5633fb25cfe2548d6+253d6470/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_64f5633fb25cfe2548d6+253d6470/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_64f5633fb25cfe2548d6+253d6470/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_64f5633fb25cfe2548d6+253d6470/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..b18f83d6d6fd3809a06c869551d7f6a94e223bb0 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_64f5633fb25cfe2548d6+253d6470/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aac8dc7e9330ce1f47f53878e2e8ce8b433677659ce2ff25966f48373cf042e3 +size 63330 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_64f5633fb25cfe2548d6+253d6470/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_64f5633fb25cfe2548d6+253d6470/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..ea1c346b87d0b2883b6cbc30fe65987dd773ec24 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_64f5633fb25cfe2548d6+253d6470/model.neff @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cae8d6499e370fa5652085a3a2d16e56022c63f8d417887d618e08d4e55bf00 +size 287744 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_78784e2fcde757cbaea0+431f5505/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_78784e2fcde757cbaea0+431f5505/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..9932c70d27d759d781be88485b74b17b49b4c04f --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_78784e2fcde757cbaea0+431f5505/compile_flags.json @@ -0,0 +1 @@ +"--model-type=transformer -O1 --lnc=1 --internal-hlo2tensorizer-options=--experimental-unsafe-fp8e4m3fn-as-fp8e4m3 --logfile=/tmp/nxd_model/layout_opt/log-neuron-cc.txt" \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_78784e2fcde757cbaea0+431f5505/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_78784e2fcde757cbaea0+431f5505/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_78784e2fcde757cbaea0+431f5505/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_78784e2fcde757cbaea0+431f5505/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..248686b0c0497ce90cbf10b915607df81e175898 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_78784e2fcde757cbaea0+431f5505/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8e300e124be19d09710fe712fd873bf2c5a1913bc08067ae5df24f2ddb9c11cf +size 9389 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_78784e2fcde757cbaea0+431f5505/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_78784e2fcde757cbaea0+431f5505/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..36a0891bb1acb70da2b3b2c0fb5405241985dcce Binary files /dev/null and b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_78784e2fcde757cbaea0+431f5505/model.neff differ diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_806ef7cec3a07d99b574+cd3419b6/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_806ef7cec3a07d99b574+cd3419b6/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..93354bd42abb66b9b6e56b127441174baa403534 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_806ef7cec3a07d99b574+cd3419b6/compile_flags.json @@ -0,0 +1 @@ +["--target=trn1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"] \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_806ef7cec3a07d99b574+cd3419b6/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_806ef7cec3a07d99b574+cd3419b6/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_806ef7cec3a07d99b574+cd3419b6/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_806ef7cec3a07d99b574+cd3419b6/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..5e48a3abb6530e6f34a6bf106fcc427a7f5b8011 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_806ef7cec3a07d99b574+cd3419b6/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2a60734c46f0e96ae9e3d939314cfa6ff355f670e23beedb9568404394687e90 +size 54868 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_806ef7cec3a07d99b574+cd3419b6/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_806ef7cec3a07d99b574+cd3419b6/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..aa513ca2c031387e945b3e01186d1e391de3d894 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_806ef7cec3a07d99b574+cd3419b6/model.neff @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a9efaba7f6635a9384cf742bb43ce1121e6daed0c080f1fe6d209783f201e372 +size 287744 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_806ef7cec3a07d99b574+cd3419b6/wrapped_neff.hlo b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_806ef7cec3a07d99b574+cd3419b6/wrapped_neff.hlo new file mode 100644 index 0000000000000000000000000000000000000000..4b38ff054b179067484adc92126c8283cfdf6a0b --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_806ef7cec3a07d99b574+cd3419b6/wrapped_neff.hlo @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:17000c8a4cfc8d65845893ec222a9efb409f4f57b06f60395073685ef946ea3d +size 297990 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_889702416621506ce9e2+cd3419b6/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_889702416621506ce9e2+cd3419b6/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..93354bd42abb66b9b6e56b127441174baa403534 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_889702416621506ce9e2+cd3419b6/compile_flags.json @@ -0,0 +1 @@ +["--target=trn1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"] \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_889702416621506ce9e2+cd3419b6/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_889702416621506ce9e2+cd3419b6/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..e29520fb2f93b45fc5210da37748e55683966461 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_889702416621506ce9e2+cd3419b6/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c6a1d6163de7397cd9023e76d6e7d1fa1bf855b31b1f0a134d60ffd529d1e99d +size 3283984 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_889702416621506ce9e2+cd3419b6/model.log b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_889702416621506ce9e2+cd3419b6/model.log new file mode 100644 index 0000000000000000000000000000000000000000..a1abb6d43198984dc74768a1c2d2ab172ab95aa7 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_889702416621506ce9e2+cd3419b6/model.log @@ -0,0 +1,3 @@ +Failed compilation with ['neuronx-cc', 'compile', '--framework=XLA', '/tmp/nxd_model/token_generation_model/_tp0_bk0/model.MODULE_889702416621506ce9e2+cd3419b6.hlo_module.pb', '--output', '/tmp/nxd_model/token_generation_model/_tp0_bk0/model.MODULE_889702416621506ce9e2+cd3419b6.neff', '--target=trn1', '--enable-saturate-infinity', '--enable-mixed-precision-accumulation', '--model-type', 'transformer', '-O1', '--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2', '--auto-cast=none', '--internal-enable-dge-levels', 'vector_dynamic_offsets', '--internal-hlo2tensorizer-options=--verify-hlo=true', '--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt', '--enable-internal-neff-wrapper', '--verbose=35']: [XCG815] Estimated peak HBM usage (27.811111) exceeds 16GB. Neff won't be able to load on chip - Please open a support ticket at https://github.com/aws-neuron/aws-neuron-sdk/issues/new. You may also be able to obtain more information using the 'XLA_IR_DEBUG' and 'XLA_HLO_DEBUG' environment variables. +2025-08-29T11:39:09Z Non-signal exit. Backend exited with code 1 and stderr: [XCG815] Estimated peak HBM usage (27.811111) exceeds 16GB. Neff won't be able to load on chip - Please open a support ticket at https://github.com/aws-neuron/aws-neuron-sdk/issues/new. You may also be able to obtain more information using the 'XLA_IR_DEBUG' and 'XLA_HLO_DEBUG' environment variables. + diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_9f17b2a2f6b76b4ae4a8+253d6470/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_9f17b2a2f6b76b4ae4a8+253d6470/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..cb53ce53583cfe21339b4bed39f5fa1a679cddd1 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_9f17b2a2f6b76b4ae4a8+253d6470/compile_flags.json @@ -0,0 +1 @@ +["--target=trn1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"] \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_9f17b2a2f6b76b4ae4a8+253d6470/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_9f17b2a2f6b76b4ae4a8+253d6470/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_9f17b2a2f6b76b4ae4a8+253d6470/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_9f17b2a2f6b76b4ae4a8+253d6470/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..07d4b550dacf6a8fcb2f3200a661d79e7d811a3c --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_9f17b2a2f6b76b4ae4a8+253d6470/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3cdf173905d4d6c94c6c80054dcca944f2f3ed60a8a91fce8fe5f9b22fc01731 +size 79738 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_9f17b2a2f6b76b4ae4a8+253d6470/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_9f17b2a2f6b76b4ae4a8+253d6470/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..08b8b766fbf9dd31ea71ffcde21fc799bbc46f2a --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_9f17b2a2f6b76b4ae4a8+253d6470/model.neff @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c477a7b110049f0649adf16dad94814fe9a8c851a79760f2aa888982ccdce658 +size 297984 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_a0dab04a6051ff2ac839+cd3419b6/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_a0dab04a6051ff2ac839+cd3419b6/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..93354bd42abb66b9b6e56b127441174baa403534 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_a0dab04a6051ff2ac839+cd3419b6/compile_flags.json @@ -0,0 +1 @@ +["--target=trn1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"] \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_a0dab04a6051ff2ac839+cd3419b6/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_a0dab04a6051ff2ac839+cd3419b6/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..28929eb9ad69274d40944b1ac44c9a66fcc4be5f --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_a0dab04a6051ff2ac839+cd3419b6/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6ba8a1147568d47fd6d9e4277affb0ec81e136d7c1d80a758a41bcb5f3e7aca9 +size 3283984 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_a0dab04a6051ff2ac839+cd3419b6/model.log b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_a0dab04a6051ff2ac839+cd3419b6/model.log new file mode 100644 index 0000000000000000000000000000000000000000..228945b6b6b0b851c0364c22476a2eb426cc57c2 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_a0dab04a6051ff2ac839+cd3419b6/model.log @@ -0,0 +1,3 @@ +Failed compilation with ['neuronx-cc', 'compile', '--framework=XLA', '/tmp/nxd_model/token_generation_model/_tp0_bk0/model.MODULE_a0dab04a6051ff2ac839+cd3419b6.hlo_module.pb', '--output', '/tmp/nxd_model/token_generation_model/_tp0_bk0/model.MODULE_a0dab04a6051ff2ac839+cd3419b6.neff', '--target=trn1', '--enable-saturate-infinity', '--enable-mixed-precision-accumulation', '--model-type', 'transformer', '-O1', '--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2', '--auto-cast=none', '--internal-enable-dge-levels', 'vector_dynamic_offsets', '--internal-hlo2tensorizer-options=--verify-hlo=true', '--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt', '--enable-internal-neff-wrapper', '--verbose=35']: [XCG815] Estimated peak HBM usage (27.672468) exceeds 16GB. Neff won't be able to load on chip - Please open a support ticket at https://github.com/aws-neuron/aws-neuron-sdk/issues/new. You may also be able to obtain more information using the 'XLA_IR_DEBUG' and 'XLA_HLO_DEBUG' environment variables. +2025-08-29T11:53:11Z Non-signal exit. Backend exited with code 1 and stderr: [XCG815] Estimated peak HBM usage (27.672468) exceeds 16GB. Neff won't be able to load on chip - Please open a support ticket at https://github.com/aws-neuron/aws-neuron-sdk/issues/new. You may also be able to obtain more information using the 'XLA_IR_DEBUG' and 'XLA_HLO_DEBUG' environment variables. + diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_ace0c273811c44dd64c1+431f5505/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_ace0c273811c44dd64c1+431f5505/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..9932c70d27d759d781be88485b74b17b49b4c04f --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_ace0c273811c44dd64c1+431f5505/compile_flags.json @@ -0,0 +1 @@ +"--model-type=transformer -O1 --lnc=1 --internal-hlo2tensorizer-options=--experimental-unsafe-fp8e4m3fn-as-fp8e4m3 --logfile=/tmp/nxd_model/layout_opt/log-neuron-cc.txt" \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_ace0c273811c44dd64c1+431f5505/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_ace0c273811c44dd64c1+431f5505/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_ace0c273811c44dd64c1+431f5505/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_ace0c273811c44dd64c1+431f5505/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..63bb4e64ed08ea18b5da1868906526e766d8bb14 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_ace0c273811c44dd64c1+431f5505/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c9ac6bfd73e125d6606fabf26ce5f60ff4043a760cc7fa9689ea7896ea7eb378 +size 10039 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_ace0c273811c44dd64c1+431f5505/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_ace0c273811c44dd64c1+431f5505/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..9f567b801b368512f309e1ab174ed398ef18413a Binary files /dev/null and b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_ace0c273811c44dd64c1+431f5505/model.neff differ diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b522f7aabb6b091bde5f+253d6470/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b522f7aabb6b091bde5f+253d6470/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..cb53ce53583cfe21339b4bed39f5fa1a679cddd1 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b522f7aabb6b091bde5f+253d6470/compile_flags.json @@ -0,0 +1 @@ +["--target=trn1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"] \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b522f7aabb6b091bde5f+253d6470/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b522f7aabb6b091bde5f+253d6470/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b522f7aabb6b091bde5f+253d6470/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b522f7aabb6b091bde5f+253d6470/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..077c105c3efa59d9c4ee34bd7139db91fe2439a0 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b522f7aabb6b091bde5f+253d6470/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:55a2643eff994253518995e9cda3f7793d7024060c5ff4de17fa34488862aec5 +size 63921 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b522f7aabb6b091bde5f+253d6470/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b522f7aabb6b091bde5f+253d6470/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..b6aed886b0f4db8d99a00be6430ecb92c3039444 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b522f7aabb6b091bde5f+253d6470/model.neff @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:400e9c32e404b3180ef75c48cbb3cd5f5b478e5cc02cd36617edaff309fd5ecd +size 267264 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b72befbd04b196887e2e+cd3419b6/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b72befbd04b196887e2e+cd3419b6/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..93354bd42abb66b9b6e56b127441174baa403534 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b72befbd04b196887e2e+cd3419b6/compile_flags.json @@ -0,0 +1 @@ +["--target=trn1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"] \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b72befbd04b196887e2e+cd3419b6/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b72befbd04b196887e2e+cd3419b6/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b72befbd04b196887e2e+cd3419b6/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b72befbd04b196887e2e+cd3419b6/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..4841d04975a14aed584672489800d2289b44dea6 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b72befbd04b196887e2e+cd3419b6/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a594a346cbcf86d47b48121fc0588a7ee7fdd8cd70ff17635b287d86ad9e350a +size 63838 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b72befbd04b196887e2e+cd3419b6/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b72befbd04b196887e2e+cd3419b6/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..c2138a20b9adf3455e07aae1462246251eefaaca --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b72befbd04b196887e2e+cd3419b6/model.neff @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:74c18cb728478c1cd597d66f35eefc0d1277ae0946c9881b8c04e1108054e3c5 +size 297984 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b72befbd04b196887e2e+cd3419b6/wrapped_neff.hlo b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b72befbd04b196887e2e+cd3419b6/wrapped_neff.hlo new file mode 100644 index 0000000000000000000000000000000000000000..59b7e8b3c0fbd09c99d122ab086b84b88517119e --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_b72befbd04b196887e2e+cd3419b6/wrapped_neff.hlo @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:142bd2d4422548fa597a1b2f80a85e0e2cb80892660b57f3ca6044c2de95f2ea +size 307570 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c02ce832872dc14d2f12+431f5505/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c02ce832872dc14d2f12+431f5505/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..9932c70d27d759d781be88485b74b17b49b4c04f --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c02ce832872dc14d2f12+431f5505/compile_flags.json @@ -0,0 +1 @@ +"--model-type=transformer -O1 --lnc=1 --internal-hlo2tensorizer-options=--experimental-unsafe-fp8e4m3fn-as-fp8e4m3 --logfile=/tmp/nxd_model/layout_opt/log-neuron-cc.txt" \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c02ce832872dc14d2f12+431f5505/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c02ce832872dc14d2f12+431f5505/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c02ce832872dc14d2f12+431f5505/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c02ce832872dc14d2f12+431f5505/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..b1cf80c06db0a363474056177a7e32484354f850 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c02ce832872dc14d2f12+431f5505/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:61c34effd5fedae7b600ec1d21f4475ef4981e5dcdb195bff8a29f8b148fd514 +size 9392 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c02ce832872dc14d2f12+431f5505/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c02ce832872dc14d2f12+431f5505/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..ce192868ec800f95235b0062d71ea90618278fe3 Binary files /dev/null and b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c02ce832872dc14d2f12+431f5505/model.neff differ diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c5daaeaa40996f359a5c+cd3419b6/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c5daaeaa40996f359a5c+cd3419b6/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..93354bd42abb66b9b6e56b127441174baa403534 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c5daaeaa40996f359a5c+cd3419b6/compile_flags.json @@ -0,0 +1 @@ +["--target=trn1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"] \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c5daaeaa40996f359a5c+cd3419b6/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c5daaeaa40996f359a5c+cd3419b6/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c5daaeaa40996f359a5c+cd3419b6/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c5daaeaa40996f359a5c+cd3419b6/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..fe8cb728bf166a8b560b8a9a7047416e88b063d5 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c5daaeaa40996f359a5c+cd3419b6/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7ef18f3eee0fea6ac1cd8e929b68e55b804dd00989b370917720d61b08b677fb +size 63933 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c5daaeaa40996f359a5c+cd3419b6/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c5daaeaa40996f359a5c+cd3419b6/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..b5d476390cb6a1fdf39bd05a78e3300291f0a6e3 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c5daaeaa40996f359a5c+cd3419b6/model.neff @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:07e125da81e6828567b56c9e7e62b90e1e63adbe89a7044321a02ac7f0d6d786 +size 297984 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c5daaeaa40996f359a5c+cd3419b6/wrapped_neff.hlo b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c5daaeaa40996f359a5c+cd3419b6/wrapped_neff.hlo new file mode 100644 index 0000000000000000000000000000000000000000..0704698365124faf9098106bec60b66337964c72 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_c5daaeaa40996f359a5c+cd3419b6/wrapped_neff.hlo @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9db505847c90dc22eec35075bdf608887e5ee6c26bf9d589da84026485cd74fa +size 306238 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_df21c2b35cb01f251b78+cd3419b6/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_df21c2b35cb01f251b78+cd3419b6/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..93354bd42abb66b9b6e56b127441174baa403534 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_df21c2b35cb01f251b78+cd3419b6/compile_flags.json @@ -0,0 +1 @@ +["--target=trn1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"] \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_df21c2b35cb01f251b78+cd3419b6/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_df21c2b35cb01f251b78+cd3419b6/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_df21c2b35cb01f251b78+cd3419b6/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_df21c2b35cb01f251b78+cd3419b6/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..d3a2ee5db2e2e28c23b4e1aa5af4a183d9ff98f9 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_df21c2b35cb01f251b78+cd3419b6/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a178d643c0c8e7e4d2209b0febfdc4cc31e7b06bbcede98c31be2a5ffc4d9f16 +size 60501 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_df21c2b35cb01f251b78+cd3419b6/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_df21c2b35cb01f251b78+cd3419b6/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..a6d4e2a619f7ec6fd038b7481572e488100c0011 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_df21c2b35cb01f251b78+cd3419b6/model.neff @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1eed0072a3c87405151a5802e5b03c5de14c06b32cd7826c08a66b7d4fa536eb +size 277504 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_df21c2b35cb01f251b78+cd3419b6/wrapped_neff.hlo b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_df21c2b35cb01f251b78+cd3419b6/wrapped_neff.hlo new file mode 100644 index 0000000000000000000000000000000000000000..977a02513f10cf1a2c2466285c36957ccfb7d87b --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_df21c2b35cb01f251b78+cd3419b6/wrapped_neff.hlo @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:68ad2ca46349cbb01096c06982a3691f5b5c2bb67eb12bd6b504922ae47878a7 +size 287750 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_f561bf7d12efdc67186d+cd3419b6/compile_flags.json b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_f561bf7d12efdc67186d+cd3419b6/compile_flags.json new file mode 100644 index 0000000000000000000000000000000000000000..93354bd42abb66b9b6e56b127441174baa403534 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_f561bf7d12efdc67186d+cd3419b6/compile_flags.json @@ -0,0 +1 @@ +["--target=trn1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"] \ No newline at end of file diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_f561bf7d12efdc67186d+cd3419b6/model.done b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_f561bf7d12efdc67186d+cd3419b6/model.done new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_f561bf7d12efdc67186d+cd3419b6/model.hlo_module.pb b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_f561bf7d12efdc67186d+cd3419b6/model.hlo_module.pb new file mode 100644 index 0000000000000000000000000000000000000000..ccd2281666b595d7154e1d2a6977ccfab28bf58d --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_f561bf7d12efdc67186d+cd3419b6/model.hlo_module.pb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:29c8356719604b8eb2bc23f160723a9bf0df82e0815e9b0112a9400d0a009629 +size 54867 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_f561bf7d12efdc67186d+cd3419b6/model.neff b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_f561bf7d12efdc67186d+cd3419b6/model.neff new file mode 100644 index 0000000000000000000000000000000000000000..4cf878c432098d5b04ebec843af45a3d3355fb8d --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_f561bf7d12efdc67186d+cd3419b6/model.neff @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4faa86504a9326e80d9e63332b3eb917b8302c62dd293d5d02fb2d5d9f729908 +size 287744 diff --git a/neuronxcc-2.19.8089.0+8ab9f450/MODULE_f561bf7d12efdc67186d+cd3419b6/wrapped_neff.hlo b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_f561bf7d12efdc67186d+cd3419b6/wrapped_neff.hlo new file mode 100644 index 0000000000000000000000000000000000000000..17f7c0c7a4349f53d203c5654d3854c895945e19 --- /dev/null +++ b/neuronxcc-2.19.8089.0+8ab9f450/MODULE_f561bf7d12efdc67186d+cd3419b6/wrapped_neff.hlo @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7b0466b7d4b1b4c175c14fc032d7155c1ae5694f0de51fd8076e94607d0ca278 +size 297990