Synchronizing local compiler cache.
Browse files- neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.0.dev5/llama/meta-llama/Meta-Llama-3.1-8B/8512e5cff891277b5d25.json +77 -0
- neuronxcc-2.19.8089.0+8ab9f450/MODULE_09f4395a4e006c5aab17+ed72d204/model.hlo_module.pb +1 -1
- neuronxcc-2.19.8089.0+8ab9f450/MODULE_09f4395a4e006c5aab17+ed72d204/model.neff +1 -1
- neuronxcc-2.19.8089.0+8ab9f450/MODULE_977a8cb77ae104c74f77+a9d440f5/model.neff +1 -1
- neuronxcc-2.19.8089.0+8ab9f450/MODULE_977a8cb77ae104c74f77+a9d440f5/wrapped_neff.hlo +1 -1
- neuronxcc-2.19.8089.0+8ab9f450/MODULE_a04897f50446b5fc1ea6+a9d440f5/model.neff +1 -1
- neuronxcc-2.19.8089.0+8ab9f450/MODULE_a04897f50446b5fc1ea6+a9d440f5/wrapped_neff.hlo +1 -1
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.0.dev5/llama/meta-llama/Meta-Llama-3.1-8B/8512e5cff891277b5d25.json
ADDED
|
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_entry_class": "SingleModelCacheEntry",
|
| 3 |
+
"_model_id": "meta-llama/Meta-Llama-3.1-8B",
|
| 4 |
+
"_task": "text-generation",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"LlamaForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"head_dim": 128,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 4096,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 14336,
|
| 15 |
+
"max_position_embeddings": 131072,
|
| 16 |
+
"mlp_bias": false,
|
| 17 |
+
"model_type": "llama",
|
| 18 |
+
"neuron": {
|
| 19 |
+
"_serialized_key": "NxDNeuronConfig",
|
| 20 |
+
"async_mode": false,
|
| 21 |
+
"attn_kernel_enabled": false,
|
| 22 |
+
"batch_size": 8,
|
| 23 |
+
"capacity_factor": null,
|
| 24 |
+
"cc_pipeline_tiling_factor": 2,
|
| 25 |
+
"checkpoint_id": "meta-llama/Meta-Llama-3.1-8B",
|
| 26 |
+
"checkpoint_revision": "d04e592bb4f6aa9cfee91e2e20afa771667e1d4b",
|
| 27 |
+
"continuous_batching": true,
|
| 28 |
+
"enable_bucketing": false,
|
| 29 |
+
"ep_degree": 1,
|
| 30 |
+
"flash_decoding_enabled": false,
|
| 31 |
+
"fused_qkv": true,
|
| 32 |
+
"glu_mlp": true,
|
| 33 |
+
"is_chunked_prefill": false,
|
| 34 |
+
"local_ranks_size": 8,
|
| 35 |
+
"logical_nc_config": 1,
|
| 36 |
+
"max_batch_size": 8,
|
| 37 |
+
"max_context_length": 4096,
|
| 38 |
+
"max_topk": 256,
|
| 39 |
+
"mlp_kernel_enabled": false,
|
| 40 |
+
"mlp_kernel_fuse_residual_add": false,
|
| 41 |
+
"n_active_tokens": 4096,
|
| 42 |
+
"neuronxcc_version": "2.19.8089.0+8ab9f450",
|
| 43 |
+
"num_cores_per_group": 1,
|
| 44 |
+
"on_device_sampling": true,
|
| 45 |
+
"optimum_neuron_version": "0.3.0.dev5",
|
| 46 |
+
"output_logits": false,
|
| 47 |
+
"padding_side": "right",
|
| 48 |
+
"pp_degree": 1,
|
| 49 |
+
"qk_layernorm": false,
|
| 50 |
+
"qkv_kernel_enabled": false,
|
| 51 |
+
"rpl_reduce_dtype": "bfloat16",
|
| 52 |
+
"sequence_length": 4096,
|
| 53 |
+
"sequence_parallel_enabled": false,
|
| 54 |
+
"speculation_length": 0,
|
| 55 |
+
"start_rank_id": 0,
|
| 56 |
+
"target": null,
|
| 57 |
+
"torch_dtype": "bfloat16",
|
| 58 |
+
"tp_degree": 8,
|
| 59 |
+
"vocab_parallel": false
|
| 60 |
+
},
|
| 61 |
+
"num_attention_heads": 32,
|
| 62 |
+
"num_hidden_layers": 32,
|
| 63 |
+
"num_key_value_heads": 8,
|
| 64 |
+
"pretraining_tp": 1,
|
| 65 |
+
"rms_norm_eps": 1e-05,
|
| 66 |
+
"rope_scaling": {
|
| 67 |
+
"factor": 8.0,
|
| 68 |
+
"high_freq_factor": 4.0,
|
| 69 |
+
"low_freq_factor": 1.0,
|
| 70 |
+
"original_max_position_embeddings": 8192,
|
| 71 |
+
"rope_type": "llama3"
|
| 72 |
+
},
|
| 73 |
+
"rope_theta": 500000.0,
|
| 74 |
+
"tie_word_embeddings": false,
|
| 75 |
+
"use_cache": true,
|
| 76 |
+
"vocab_size": 128256
|
| 77 |
+
}
|
neuronxcc-2.19.8089.0+8ab9f450/MODULE_09f4395a4e006c5aab17+ed72d204/model.hlo_module.pb
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 947655
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ce8ab69c78f9087ad35e56c2f49ab70442de201a04a36e6c0557177a7165ec71
|
| 3 |
size 947655
|
neuronxcc-2.19.8089.0+8ab9f450/MODULE_09f4395a4e006c5aab17+ed72d204/model.neff
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 31130624
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:206f0245d8735f1169a6a8bd482d3e9bc577903f4e9634beaa8c01b4af168750
|
| 3 |
size 31130624
|
neuronxcc-2.19.8089.0+8ab9f450/MODULE_977a8cb77ae104c74f77+a9d440f5/model.neff
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3144704
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1b137e3e7e9461d21dae4b107a9d66542860e3ec5f0d9766890aa2c705fef9e1
|
| 3 |
size 3144704
|
neuronxcc-2.19.8089.0+8ab9f450/MODULE_977a8cb77ae104c74f77+a9d440f5/wrapped_neff.hlo
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3282521
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:38c4faa0a5dcd29db15650b3f433c30d0743e07ea21103363a9efd6074fbb77c
|
| 3 |
size 3282521
|
neuronxcc-2.19.8089.0+8ab9f450/MODULE_a04897f50446b5fc1ea6+a9d440f5/model.neff
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2509824
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3b6724b129b0e15a1e673b51eb5d54aeff8e3144f8454ff38d0d1e9db9b51af9
|
| 3 |
size 2509824
|
neuronxcc-2.19.8089.0+8ab9f450/MODULE_a04897f50446b5fc1ea6+a9d440f5/wrapped_neff.hlo
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2647641
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2c525fea5f0439318b60a4000ca1900953b4067372cfa9d227b9cbf68a48be01
|
| 3 |
size 2647641
|