diff --git a/added_tokens.json b/added_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..56f8f607add57bb7f9c2a5cb5c57865e1fcf2b31
--- /dev/null
+++ b/added_tokens.json
@@ -0,0 +1,9 @@
+{
+ "<|im_end|>": 122753,
+ "<|im_start|>": 122757,
+ "<|tool_call|>": 122756,
+ "▁": 122758,
+ "▁": 122755,
+ "▁": 122754,
+ "▁": 122759
+}
diff --git a/mlc-chat-config.json b/mlc-chat-config.json
new file mode 100644
index 0000000000000000000000000000000000000000..18e0324d4a254844c6972a035658fd97a28bc768
--- /dev/null
+++ b/mlc-chat-config.json
@@ -0,0 +1,86 @@
+{
+ "version": "0.1.0",
+ "model_type": "minicpm",
+ "quantization": "q4f16_1",
+ "model_config": {
+ "vocab_size": 122760,
+ "hidden_size": 2304,
+ "num_hidden_layers": 40,
+ "num_attention_heads": 36,
+ "num_key_value_heads": 36,
+ "hidden_act": "silu",
+ "rms_norm_eps": 1e-05,
+ "intermediate_size": 5760,
+ "scale_emb": 12,
+ "scale_depth": 1.4,
+ "dim_model_base": 256,
+ "use_cache": true,
+ "bos_token_id": 1,
+ "eos_token_id": 2,
+ "tie_word_embeddings": false,
+ "rope_theta": 1000000.0,
+ "context_window_size": 65536,
+ "prefill_chunk_size": 8192,
+ "tensor_parallel_shards": 1,
+ "head_dim": 64,
+ "max_batch_size": 128,
+ "num_experts_per_tok": 0,
+ "num_experts": 0
+ },
+ "vocab_size": 122760,
+ "context_window_size": 65536,
+ "sliding_window_size": -1,
+ "prefill_chunk_size": 8192,
+ "attention_sink_size": -1,
+ "tensor_parallel_shards": 1,
+ "pipeline_parallel_stages": 1,
+ "temperature": 1.0,
+ "presence_penalty": 0.0,
+ "frequency_penalty": 0.0,
+ "repetition_penalty": 1.0,
+ "top_p": 1.0,
+ "tokenizer_files": [
+ "tokenizer.model",
+ "tokenizer.json",
+ "added_tokens.json",
+ "tokenizer_config.json"
+ ],
+ "tokenizer_info": {
+ "token_postproc_method": "byte_fallback",
+ "prepend_space_in_encode": true,
+ "strip_space_in_decode": true
+ },
+ "conv_template": {
+ "name": "chatml",
+ "system_template": "<|im_start|>system\n{system_message}<|im_end|>\n",
+ "system_message": "A conversation between a user and an LLM-based AI assistant. The assistant gives helpful and honest answers.",
+ "system_prefix_token_ids": null,
+ "add_role_after_system_message": true,
+ "roles": {
+ "user": "<|im_start|>user",
+ "assistant": "<|im_start|>assistant"
+ },
+ "role_templates": {
+ "user": "{user_message}",
+ "assistant": "{assistant_message}",
+ "tool": "{tool_message}"
+ },
+ "messages": [],
+ "seps": [
+ "<|im_end|>\n"
+ ],
+ "role_content_sep": "\n",
+ "role_empty_sep": "\n",
+ "stop_str": [
+ "<|im_end|>"
+ ],
+ "stop_token_ids": [
+ 2
+ ],
+ "function_string": "",
+ "use_function_calling": false
+ },
+ "pad_token_id": 0,
+ "bos_token_id": 1,
+ "eos_token_id": 2
+}
\ No newline at end of file
diff --git a/ndarray-cache.json b/ndarray-cache.json
new file mode 100644
index 0000000000000000000000000000000000000000..d6c4ada44ab80ad62ebf5ac83bc91e7d21d098ef
--- /dev/null
+++ b/ndarray-cache.json
@@ -0,0 +1,4791 @@
+{
+ "metadata": {
+ "ParamSize": 405,
+ "ParamBytes": 1692119808.0,
+ "BitsPerParam": 4.500713551972092
+ },
+ "records": [
+ {
+ "dataPath": "params_shard_0.bin",
+ "format": "raw-shard",
+ "nbytes": 141419520,
+ "records": [
+ {
+ "name": "model.embed_tokens.q_weight",
+ "shape": [
+ 122760,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 141419520,
+ "byteOffset": 0
+ }
+ ],
+ "md5sum": "673a7768cedf15f588c8982fbb19a2a4"
+ },
+ {
+ "dataPath": "params_shard_1.bin",
+ "format": "raw-shard",
+ "nbytes": 141419520,
+ "records": [
+ {
+ "name": "lm_head.q_weight",
+ "shape": [
+ 122760,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 141419520,
+ "byteOffset": 0
+ }
+ ],
+ "md5sum": "f905741e89ce94cf94c67ea8dd3065ac"
+ },
+ {
+ "dataPath": "params_shard_2.bin",
+ "format": "raw-shard",
+ "nbytes": 17677440,
+ "records": [
+ {
+ "name": "lm_head.q_scale",
+ "shape": [
+ 122760,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 17677440,
+ "byteOffset": 0
+ }
+ ],
+ "md5sum": "3877afae3c12668e8b7d093b4d8ad572"
+ },
+ {
+ "dataPath": "params_shard_3.bin",
+ "format": "raw-shard",
+ "nbytes": 29635200,
+ "records": [
+ {
+ "name": "model.embed_tokens.q_scale",
+ "shape": [
+ 122760,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 17677440,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.norm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 17677440
+ },
+ {
+ "name": "model.layers.0.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 17682048
+ },
+ {
+ "name": "model.layers.0.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 17686656
+ },
+ {
+ "name": "model.layers.0.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 25649280
+ },
+ {
+ "name": "model.layers.0.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 26644608
+ },
+ {
+ "name": "model.layers.0.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 29298816
+ },
+ {
+ "name": "model.layers.0.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 29630592
+ }
+ ],
+ "md5sum": "085ef960e2f6c8132e3fb444a605e452"
+ },
+ {
+ "dataPath": "params_shard_4.bin",
+ "format": "raw-shard",
+ "nbytes": 31357440,
+ "records": [
+ {
+ "name": "model.layers.0.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.0.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 13271040
+ },
+ {
+ "name": "model.layers.0.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 14929920
+ },
+ {
+ "name": "model.layers.0.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 21565440
+ },
+ {
+ "name": "model.layers.1.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 22394880
+ },
+ {
+ "name": "model.layers.1.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 22399488
+ },
+ {
+ "name": "model.layers.1.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 30362112
+ }
+ ],
+ "md5sum": "022837b6bdaff3659897cfe5d78537c9"
+ },
+ {
+ "dataPath": "params_shard_5.bin",
+ "format": "raw-shard",
+ "nbytes": 33352704,
+ "records": [
+ {
+ "name": "model.layers.1.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.1.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 2654208
+ },
+ {
+ "name": "model.layers.1.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 2985984
+ },
+ {
+ "name": "model.layers.1.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 2990592
+ },
+ {
+ "name": "model.layers.1.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 16261632
+ },
+ {
+ "name": "model.layers.1.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 17920512
+ },
+ {
+ "name": "model.layers.1.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 24556032
+ },
+ {
+ "name": "model.layers.2.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 25385472
+ },
+ {
+ "name": "model.layers.2.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 25390080
+ }
+ ],
+ "md5sum": "c27346f8415db11e1da2c9243c4f21fa"
+ },
+ {
+ "dataPath": "params_shard_6.bin",
+ "format": "raw-shard",
+ "nbytes": 26385408,
+ "records": [
+ {
+ "name": "model.layers.2.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.2.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 995328
+ },
+ {
+ "name": "model.layers.2.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 3649536
+ },
+ {
+ "name": "model.layers.2.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 3981312
+ },
+ {
+ "name": "model.layers.2.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 3985920
+ },
+ {
+ "name": "model.layers.2.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 17256960
+ },
+ {
+ "name": "model.layers.2.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 18915840
+ },
+ {
+ "name": "model.layers.2.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 25551360
+ },
+ {
+ "name": "model.layers.3.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 26380800
+ }
+ ],
+ "md5sum": "5583a2127c66d85de1337f2f09fbe8e5"
+ },
+ {
+ "dataPath": "params_shard_7.bin",
+ "format": "raw-shard",
+ "nbytes": 33513984,
+ "records": [
+ {
+ "name": "model.layers.3.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.3.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 7962624
+ },
+ {
+ "name": "model.layers.3.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 8957952
+ },
+ {
+ "name": "model.layers.3.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 11612160
+ },
+ {
+ "name": "model.layers.3.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 11943936
+ },
+ {
+ "name": "model.layers.3.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 11948544
+ },
+ {
+ "name": "model.layers.3.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 25219584
+ },
+ {
+ "name": "model.layers.3.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 26878464
+ }
+ ],
+ "md5sum": "f5413b64b04950c996f6702c22853185"
+ },
+ {
+ "dataPath": "params_shard_8.bin",
+ "format": "raw-shard",
+ "nbytes": 27712512,
+ "records": [
+ {
+ "name": "model.layers.3.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.4.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 829440
+ },
+ {
+ "name": "model.layers.4.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 834048
+ },
+ {
+ "name": "model.layers.4.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 8796672
+ },
+ {
+ "name": "model.layers.4.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 9792000
+ },
+ {
+ "name": "model.layers.4.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 12446208
+ },
+ {
+ "name": "model.layers.4.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 12777984
+ },
+ {
+ "name": "model.layers.4.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 12782592
+ },
+ {
+ "name": "model.layers.4.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 26053632
+ }
+ ],
+ "md5sum": "7b76244f6188c18620eb4727b5ff2ed8"
+ },
+ {
+ "dataPath": "params_shard_9.bin",
+ "format": "raw-shard",
+ "nbytes": 32689152,
+ "records": [
+ {
+ "name": "model.layers.4.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.4.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 6635520
+ },
+ {
+ "name": "model.layers.5.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 7464960
+ },
+ {
+ "name": "model.layers.5.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 7469568
+ },
+ {
+ "name": "model.layers.5.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 15432192
+ },
+ {
+ "name": "model.layers.5.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 16427520
+ },
+ {
+ "name": "model.layers.5.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 19081728
+ },
+ {
+ "name": "model.layers.5.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 19413504
+ },
+ {
+ "name": "model.layers.5.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 19418112
+ }
+ ],
+ "md5sum": "b7d58215dabb597121af7dd961f6f84c"
+ },
+ {
+ "dataPath": "params_shard_10.bin",
+ "format": "raw-shard",
+ "nbytes": 21076992,
+ "records": [
+ {
+ "name": "model.layers.5.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.5.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 1658880
+ },
+ {
+ "name": "model.layers.5.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 8294400
+ },
+ {
+ "name": "model.layers.6.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 9123840
+ },
+ {
+ "name": "model.layers.6.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 9128448
+ },
+ {
+ "name": "model.layers.6.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 17091072
+ },
+ {
+ "name": "model.layers.6.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 18086400
+ },
+ {
+ "name": "model.layers.6.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 20740608
+ },
+ {
+ "name": "model.layers.6.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 21072384
+ }
+ ],
+ "md5sum": "d185995620198a7c2ed217942d22a90f"
+ },
+ {
+ "dataPath": "params_shard_11.bin",
+ "format": "raw-shard",
+ "nbytes": 31357440,
+ "records": [
+ {
+ "name": "model.layers.6.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.6.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 13271040
+ },
+ {
+ "name": "model.layers.6.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 14929920
+ },
+ {
+ "name": "model.layers.6.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 21565440
+ },
+ {
+ "name": "model.layers.7.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 22394880
+ },
+ {
+ "name": "model.layers.7.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 22399488
+ },
+ {
+ "name": "model.layers.7.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 30362112
+ }
+ ],
+ "md5sum": "7ae1d3896da91089c113f0cdd5a16c40"
+ },
+ {
+ "dataPath": "params_shard_12.bin",
+ "format": "raw-shard",
+ "nbytes": 33352704,
+ "records": [
+ {
+ "name": "model.layers.7.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.7.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 2654208
+ },
+ {
+ "name": "model.layers.7.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 2985984
+ },
+ {
+ "name": "model.layers.7.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 2990592
+ },
+ {
+ "name": "model.layers.7.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 16261632
+ },
+ {
+ "name": "model.layers.7.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 17920512
+ },
+ {
+ "name": "model.layers.7.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 24556032
+ },
+ {
+ "name": "model.layers.8.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 25385472
+ },
+ {
+ "name": "model.layers.8.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 25390080
+ }
+ ],
+ "md5sum": "48f2f038307f162624ece8751f7a77e1"
+ },
+ {
+ "dataPath": "params_shard_13.bin",
+ "format": "raw-shard",
+ "nbytes": 26385408,
+ "records": [
+ {
+ "name": "model.layers.8.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.8.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 995328
+ },
+ {
+ "name": "model.layers.8.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 3649536
+ },
+ {
+ "name": "model.layers.8.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 3981312
+ },
+ {
+ "name": "model.layers.8.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 3985920
+ },
+ {
+ "name": "model.layers.8.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 17256960
+ },
+ {
+ "name": "model.layers.8.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 18915840
+ },
+ {
+ "name": "model.layers.8.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 25551360
+ },
+ {
+ "name": "model.layers.9.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 26380800
+ }
+ ],
+ "md5sum": "9812e52a470f3b6eb23d46669395e251"
+ },
+ {
+ "dataPath": "params_shard_14.bin",
+ "format": "raw-shard",
+ "nbytes": 33513984,
+ "records": [
+ {
+ "name": "model.layers.9.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.9.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 7962624
+ },
+ {
+ "name": "model.layers.9.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 8957952
+ },
+ {
+ "name": "model.layers.9.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 11612160
+ },
+ {
+ "name": "model.layers.9.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 11943936
+ },
+ {
+ "name": "model.layers.9.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 11948544
+ },
+ {
+ "name": "model.layers.9.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 25219584
+ },
+ {
+ "name": "model.layers.9.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 26878464
+ }
+ ],
+ "md5sum": "bccced049950d692c8971a2497170bde"
+ },
+ {
+ "dataPath": "params_shard_15.bin",
+ "format": "raw-shard",
+ "nbytes": 27712512,
+ "records": [
+ {
+ "name": "model.layers.9.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.10.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 829440
+ },
+ {
+ "name": "model.layers.10.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 834048
+ },
+ {
+ "name": "model.layers.10.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 8796672
+ },
+ {
+ "name": "model.layers.10.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 9792000
+ },
+ {
+ "name": "model.layers.10.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 12446208
+ },
+ {
+ "name": "model.layers.10.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 12777984
+ },
+ {
+ "name": "model.layers.10.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 12782592
+ },
+ {
+ "name": "model.layers.10.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 26053632
+ }
+ ],
+ "md5sum": "edcc023bef5daa1ba8b8026abf91dde3"
+ },
+ {
+ "dataPath": "params_shard_16.bin",
+ "format": "raw-shard",
+ "nbytes": 32689152,
+ "records": [
+ {
+ "name": "model.layers.10.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.10.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 6635520
+ },
+ {
+ "name": "model.layers.11.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 7464960
+ },
+ {
+ "name": "model.layers.11.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 7469568
+ },
+ {
+ "name": "model.layers.11.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 15432192
+ },
+ {
+ "name": "model.layers.11.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 16427520
+ },
+ {
+ "name": "model.layers.11.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 19081728
+ },
+ {
+ "name": "model.layers.11.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 19413504
+ },
+ {
+ "name": "model.layers.11.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 19418112
+ }
+ ],
+ "md5sum": "e61459ee617321735c15b7b0abb03bad"
+ },
+ {
+ "dataPath": "params_shard_17.bin",
+ "format": "raw-shard",
+ "nbytes": 21076992,
+ "records": [
+ {
+ "name": "model.layers.11.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.11.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 1658880
+ },
+ {
+ "name": "model.layers.11.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 8294400
+ },
+ {
+ "name": "model.layers.12.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 9123840
+ },
+ {
+ "name": "model.layers.12.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 9128448
+ },
+ {
+ "name": "model.layers.12.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 17091072
+ },
+ {
+ "name": "model.layers.12.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 18086400
+ },
+ {
+ "name": "model.layers.12.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 20740608
+ },
+ {
+ "name": "model.layers.12.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 21072384
+ }
+ ],
+ "md5sum": "fb932d92393c5478da72034a79c3e17f"
+ },
+ {
+ "dataPath": "params_shard_18.bin",
+ "format": "raw-shard",
+ "nbytes": 31357440,
+ "records": [
+ {
+ "name": "model.layers.12.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.12.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 13271040
+ },
+ {
+ "name": "model.layers.12.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 14929920
+ },
+ {
+ "name": "model.layers.12.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 21565440
+ },
+ {
+ "name": "model.layers.13.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 22394880
+ },
+ {
+ "name": "model.layers.13.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 22399488
+ },
+ {
+ "name": "model.layers.13.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 30362112
+ }
+ ],
+ "md5sum": "61c8c8f37ab150eabeebfc1639361b7b"
+ },
+ {
+ "dataPath": "params_shard_19.bin",
+ "format": "raw-shard",
+ "nbytes": 33352704,
+ "records": [
+ {
+ "name": "model.layers.13.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.13.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 2654208
+ },
+ {
+ "name": "model.layers.13.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 2985984
+ },
+ {
+ "name": "model.layers.13.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 2990592
+ },
+ {
+ "name": "model.layers.13.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 16261632
+ },
+ {
+ "name": "model.layers.13.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 17920512
+ },
+ {
+ "name": "model.layers.13.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 24556032
+ },
+ {
+ "name": "model.layers.14.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 25385472
+ },
+ {
+ "name": "model.layers.14.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 25390080
+ }
+ ],
+ "md5sum": "47f24b91b27a5e5a684130e13c690d47"
+ },
+ {
+ "dataPath": "params_shard_20.bin",
+ "format": "raw-shard",
+ "nbytes": 26385408,
+ "records": [
+ {
+ "name": "model.layers.14.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.14.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 995328
+ },
+ {
+ "name": "model.layers.14.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 3649536
+ },
+ {
+ "name": "model.layers.14.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 3981312
+ },
+ {
+ "name": "model.layers.14.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 3985920
+ },
+ {
+ "name": "model.layers.14.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 17256960
+ },
+ {
+ "name": "model.layers.14.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 18915840
+ },
+ {
+ "name": "model.layers.14.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 25551360
+ },
+ {
+ "name": "model.layers.15.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 26380800
+ }
+ ],
+ "md5sum": "cdfa99bbb314239b6247ea6ea7063565"
+ },
+ {
+ "dataPath": "params_shard_21.bin",
+ "format": "raw-shard",
+ "nbytes": 33513984,
+ "records": [
+ {
+ "name": "model.layers.15.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.15.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 7962624
+ },
+ {
+ "name": "model.layers.15.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 8957952
+ },
+ {
+ "name": "model.layers.15.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 11612160
+ },
+ {
+ "name": "model.layers.15.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 11943936
+ },
+ {
+ "name": "model.layers.15.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 11948544
+ },
+ {
+ "name": "model.layers.15.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 25219584
+ },
+ {
+ "name": "model.layers.15.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 26878464
+ }
+ ],
+ "md5sum": "436fba893a097c52533be5b5409522ba"
+ },
+ {
+ "dataPath": "params_shard_22.bin",
+ "format": "raw-shard",
+ "nbytes": 27712512,
+ "records": [
+ {
+ "name": "model.layers.15.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.16.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 829440
+ },
+ {
+ "name": "model.layers.16.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 834048
+ },
+ {
+ "name": "model.layers.16.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 8796672
+ },
+ {
+ "name": "model.layers.16.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 9792000
+ },
+ {
+ "name": "model.layers.16.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 12446208
+ },
+ {
+ "name": "model.layers.16.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 12777984
+ },
+ {
+ "name": "model.layers.16.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 12782592
+ },
+ {
+ "name": "model.layers.16.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 26053632
+ }
+ ],
+ "md5sum": "8512f451b87f7f62fe3b5602ca0b8143"
+ },
+ {
+ "dataPath": "params_shard_23.bin",
+ "format": "raw-shard",
+ "nbytes": 32689152,
+ "records": [
+ {
+ "name": "model.layers.16.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.16.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 6635520
+ },
+ {
+ "name": "model.layers.17.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 7464960
+ },
+ {
+ "name": "model.layers.17.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 7469568
+ },
+ {
+ "name": "model.layers.17.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 15432192
+ },
+ {
+ "name": "model.layers.17.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 16427520
+ },
+ {
+ "name": "model.layers.17.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 19081728
+ },
+ {
+ "name": "model.layers.17.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 19413504
+ },
+ {
+ "name": "model.layers.17.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 19418112
+ }
+ ],
+ "md5sum": "8eb937c7723d2dcc0bdfeca4f5ec0064"
+ },
+ {
+ "dataPath": "params_shard_24.bin",
+ "format": "raw-shard",
+ "nbytes": 21076992,
+ "records": [
+ {
+ "name": "model.layers.17.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.17.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 1658880
+ },
+ {
+ "name": "model.layers.17.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 8294400
+ },
+ {
+ "name": "model.layers.18.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 9123840
+ },
+ {
+ "name": "model.layers.18.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 9128448
+ },
+ {
+ "name": "model.layers.18.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 17091072
+ },
+ {
+ "name": "model.layers.18.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 18086400
+ },
+ {
+ "name": "model.layers.18.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 20740608
+ },
+ {
+ "name": "model.layers.18.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 21072384
+ }
+ ],
+ "md5sum": "d268a1e5e2134681469776369e545ee7"
+ },
+ {
+ "dataPath": "params_shard_25.bin",
+ "format": "raw-shard",
+ "nbytes": 31357440,
+ "records": [
+ {
+ "name": "model.layers.18.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.18.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 13271040
+ },
+ {
+ "name": "model.layers.18.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 14929920
+ },
+ {
+ "name": "model.layers.18.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 21565440
+ },
+ {
+ "name": "model.layers.19.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 22394880
+ },
+ {
+ "name": "model.layers.19.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 22399488
+ },
+ {
+ "name": "model.layers.19.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 30362112
+ }
+ ],
+ "md5sum": "acd08cfd0a3e23c47392323f6ceff52b"
+ },
+ {
+ "dataPath": "params_shard_26.bin",
+ "format": "raw-shard",
+ "nbytes": 33352704,
+ "records": [
+ {
+ "name": "model.layers.19.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.19.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 2654208
+ },
+ {
+ "name": "model.layers.19.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 2985984
+ },
+ {
+ "name": "model.layers.19.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 2990592
+ },
+ {
+ "name": "model.layers.19.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 16261632
+ },
+ {
+ "name": "model.layers.19.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 17920512
+ },
+ {
+ "name": "model.layers.19.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 24556032
+ },
+ {
+ "name": "model.layers.20.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 25385472
+ },
+ {
+ "name": "model.layers.20.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 25390080
+ }
+ ],
+ "md5sum": "4d521c96873595e62ca348825935f929"
+ },
+ {
+ "dataPath": "params_shard_27.bin",
+ "format": "raw-shard",
+ "nbytes": 26385408,
+ "records": [
+ {
+ "name": "model.layers.20.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.20.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 995328
+ },
+ {
+ "name": "model.layers.20.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 3649536
+ },
+ {
+ "name": "model.layers.20.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 3981312
+ },
+ {
+ "name": "model.layers.20.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 3985920
+ },
+ {
+ "name": "model.layers.20.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 17256960
+ },
+ {
+ "name": "model.layers.20.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 18915840
+ },
+ {
+ "name": "model.layers.20.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 25551360
+ },
+ {
+ "name": "model.layers.21.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 26380800
+ }
+ ],
+ "md5sum": "60648647d3c34883231d031a06d8a327"
+ },
+ {
+ "dataPath": "params_shard_28.bin",
+ "format": "raw-shard",
+ "nbytes": 33513984,
+ "records": [
+ {
+ "name": "model.layers.21.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.21.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 7962624
+ },
+ {
+ "name": "model.layers.21.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 8957952
+ },
+ {
+ "name": "model.layers.21.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 11612160
+ },
+ {
+ "name": "model.layers.21.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 11943936
+ },
+ {
+ "name": "model.layers.21.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 11948544
+ },
+ {
+ "name": "model.layers.21.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 25219584
+ },
+ {
+ "name": "model.layers.21.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 26878464
+ }
+ ],
+ "md5sum": "4d1d2bfa37467fdbad72640fd3c4392d"
+ },
+ {
+ "dataPath": "params_shard_29.bin",
+ "format": "raw-shard",
+ "nbytes": 27712512,
+ "records": [
+ {
+ "name": "model.layers.21.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.22.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 829440
+ },
+ {
+ "name": "model.layers.22.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 834048
+ },
+ {
+ "name": "model.layers.22.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 8796672
+ },
+ {
+ "name": "model.layers.22.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 9792000
+ },
+ {
+ "name": "model.layers.22.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 12446208
+ },
+ {
+ "name": "model.layers.22.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 12777984
+ },
+ {
+ "name": "model.layers.22.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 12782592
+ },
+ {
+ "name": "model.layers.22.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 26053632
+ }
+ ],
+ "md5sum": "b106bab4f05c51001c5be6939f7936df"
+ },
+ {
+ "dataPath": "params_shard_30.bin",
+ "format": "raw-shard",
+ "nbytes": 32689152,
+ "records": [
+ {
+ "name": "model.layers.22.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.22.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 6635520
+ },
+ {
+ "name": "model.layers.23.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 7464960
+ },
+ {
+ "name": "model.layers.23.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 7469568
+ },
+ {
+ "name": "model.layers.23.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 15432192
+ },
+ {
+ "name": "model.layers.23.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 16427520
+ },
+ {
+ "name": "model.layers.23.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 19081728
+ },
+ {
+ "name": "model.layers.23.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 19413504
+ },
+ {
+ "name": "model.layers.23.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 19418112
+ }
+ ],
+ "md5sum": "8dff849a1597e03193110c587588f28f"
+ },
+ {
+ "dataPath": "params_shard_31.bin",
+ "format": "raw-shard",
+ "nbytes": 21076992,
+ "records": [
+ {
+ "name": "model.layers.23.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.23.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 1658880
+ },
+ {
+ "name": "model.layers.23.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 8294400
+ },
+ {
+ "name": "model.layers.24.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 9123840
+ },
+ {
+ "name": "model.layers.24.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 9128448
+ },
+ {
+ "name": "model.layers.24.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 17091072
+ },
+ {
+ "name": "model.layers.24.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 18086400
+ },
+ {
+ "name": "model.layers.24.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 20740608
+ },
+ {
+ "name": "model.layers.24.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 21072384
+ }
+ ],
+ "md5sum": "ffe49e22b525b2ca9c422e930863fd6d"
+ },
+ {
+ "dataPath": "params_shard_32.bin",
+ "format": "raw-shard",
+ "nbytes": 31357440,
+ "records": [
+ {
+ "name": "model.layers.24.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.24.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 13271040
+ },
+ {
+ "name": "model.layers.24.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 14929920
+ },
+ {
+ "name": "model.layers.24.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 21565440
+ },
+ {
+ "name": "model.layers.25.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 22394880
+ },
+ {
+ "name": "model.layers.25.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 22399488
+ },
+ {
+ "name": "model.layers.25.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 30362112
+ }
+ ],
+ "md5sum": "819339c3705965b5cc4ffeba9515531d"
+ },
+ {
+ "dataPath": "params_shard_33.bin",
+ "format": "raw-shard",
+ "nbytes": 33352704,
+ "records": [
+ {
+ "name": "model.layers.25.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.25.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 2654208
+ },
+ {
+ "name": "model.layers.25.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 2985984
+ },
+ {
+ "name": "model.layers.25.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 2990592
+ },
+ {
+ "name": "model.layers.25.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 16261632
+ },
+ {
+ "name": "model.layers.25.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 17920512
+ },
+ {
+ "name": "model.layers.25.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 24556032
+ },
+ {
+ "name": "model.layers.26.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 25385472
+ },
+ {
+ "name": "model.layers.26.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 25390080
+ }
+ ],
+ "md5sum": "246505c91eb3c83919d381147143ce42"
+ },
+ {
+ "dataPath": "params_shard_34.bin",
+ "format": "raw-shard",
+ "nbytes": 26385408,
+ "records": [
+ {
+ "name": "model.layers.26.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.26.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 995328
+ },
+ {
+ "name": "model.layers.26.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 3649536
+ },
+ {
+ "name": "model.layers.26.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 3981312
+ },
+ {
+ "name": "model.layers.26.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 3985920
+ },
+ {
+ "name": "model.layers.26.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 17256960
+ },
+ {
+ "name": "model.layers.26.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 18915840
+ },
+ {
+ "name": "model.layers.26.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 25551360
+ },
+ {
+ "name": "model.layers.27.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 26380800
+ }
+ ],
+ "md5sum": "d6bea613808438798ac157eb503ba707"
+ },
+ {
+ "dataPath": "params_shard_35.bin",
+ "format": "raw-shard",
+ "nbytes": 33513984,
+ "records": [
+ {
+ "name": "model.layers.27.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.27.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 7962624
+ },
+ {
+ "name": "model.layers.27.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 8957952
+ },
+ {
+ "name": "model.layers.27.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 11612160
+ },
+ {
+ "name": "model.layers.27.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 11943936
+ },
+ {
+ "name": "model.layers.27.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 11948544
+ },
+ {
+ "name": "model.layers.27.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 25219584
+ },
+ {
+ "name": "model.layers.27.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 26878464
+ }
+ ],
+ "md5sum": "6e9eec8ba391133e276a4c26d16ac1f7"
+ },
+ {
+ "dataPath": "params_shard_36.bin",
+ "format": "raw-shard",
+ "nbytes": 27712512,
+ "records": [
+ {
+ "name": "model.layers.27.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.28.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 829440
+ },
+ {
+ "name": "model.layers.28.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 834048
+ },
+ {
+ "name": "model.layers.28.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 8796672
+ },
+ {
+ "name": "model.layers.28.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 9792000
+ },
+ {
+ "name": "model.layers.28.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 12446208
+ },
+ {
+ "name": "model.layers.28.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 12777984
+ },
+ {
+ "name": "model.layers.28.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 12782592
+ },
+ {
+ "name": "model.layers.28.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 26053632
+ }
+ ],
+ "md5sum": "14857aaa301e493b9dc7b95964d8b2f3"
+ },
+ {
+ "dataPath": "params_shard_37.bin",
+ "format": "raw-shard",
+ "nbytes": 32689152,
+ "records": [
+ {
+ "name": "model.layers.28.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.28.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 6635520
+ },
+ {
+ "name": "model.layers.29.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 7464960
+ },
+ {
+ "name": "model.layers.29.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 7469568
+ },
+ {
+ "name": "model.layers.29.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 15432192
+ },
+ {
+ "name": "model.layers.29.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 16427520
+ },
+ {
+ "name": "model.layers.29.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 19081728
+ },
+ {
+ "name": "model.layers.29.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 19413504
+ },
+ {
+ "name": "model.layers.29.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 19418112
+ }
+ ],
+ "md5sum": "32a11753fd3352de32a58a034ca84216"
+ },
+ {
+ "dataPath": "params_shard_38.bin",
+ "format": "raw-shard",
+ "nbytes": 21076992,
+ "records": [
+ {
+ "name": "model.layers.29.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.29.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 1658880
+ },
+ {
+ "name": "model.layers.29.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 8294400
+ },
+ {
+ "name": "model.layers.30.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 9123840
+ },
+ {
+ "name": "model.layers.30.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 9128448
+ },
+ {
+ "name": "model.layers.30.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 17091072
+ },
+ {
+ "name": "model.layers.30.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 18086400
+ },
+ {
+ "name": "model.layers.30.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 20740608
+ },
+ {
+ "name": "model.layers.30.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 21072384
+ }
+ ],
+ "md5sum": "c5310893e62902b8bcd809641d477066"
+ },
+ {
+ "dataPath": "params_shard_39.bin",
+ "format": "raw-shard",
+ "nbytes": 31357440,
+ "records": [
+ {
+ "name": "model.layers.30.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.30.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 13271040
+ },
+ {
+ "name": "model.layers.30.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 14929920
+ },
+ {
+ "name": "model.layers.30.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 21565440
+ },
+ {
+ "name": "model.layers.31.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 22394880
+ },
+ {
+ "name": "model.layers.31.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 22399488
+ },
+ {
+ "name": "model.layers.31.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 30362112
+ }
+ ],
+ "md5sum": "1978547a205de5b5ed1db9b30317048e"
+ },
+ {
+ "dataPath": "params_shard_40.bin",
+ "format": "raw-shard",
+ "nbytes": 33352704,
+ "records": [
+ {
+ "name": "model.layers.31.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.31.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 2654208
+ },
+ {
+ "name": "model.layers.31.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 2985984
+ },
+ {
+ "name": "model.layers.31.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 2990592
+ },
+ {
+ "name": "model.layers.31.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 16261632
+ },
+ {
+ "name": "model.layers.31.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 17920512
+ },
+ {
+ "name": "model.layers.31.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 24556032
+ },
+ {
+ "name": "model.layers.32.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 25385472
+ },
+ {
+ "name": "model.layers.32.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 25390080
+ }
+ ],
+ "md5sum": "8083282bca9b01b1a42d92ee6dade96f"
+ },
+ {
+ "dataPath": "params_shard_41.bin",
+ "format": "raw-shard",
+ "nbytes": 26385408,
+ "records": [
+ {
+ "name": "model.layers.32.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.32.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 995328
+ },
+ {
+ "name": "model.layers.32.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 3649536
+ },
+ {
+ "name": "model.layers.32.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 3981312
+ },
+ {
+ "name": "model.layers.32.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 3985920
+ },
+ {
+ "name": "model.layers.32.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 17256960
+ },
+ {
+ "name": "model.layers.32.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 18915840
+ },
+ {
+ "name": "model.layers.32.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 25551360
+ },
+ {
+ "name": "model.layers.33.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 26380800
+ }
+ ],
+ "md5sum": "6c371e43d8e96ca8fa0246002fff3085"
+ },
+ {
+ "dataPath": "params_shard_42.bin",
+ "format": "raw-shard",
+ "nbytes": 33513984,
+ "records": [
+ {
+ "name": "model.layers.33.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.33.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 7962624
+ },
+ {
+ "name": "model.layers.33.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 8957952
+ },
+ {
+ "name": "model.layers.33.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 11612160
+ },
+ {
+ "name": "model.layers.33.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 11943936
+ },
+ {
+ "name": "model.layers.33.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 11948544
+ },
+ {
+ "name": "model.layers.33.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 25219584
+ },
+ {
+ "name": "model.layers.33.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 26878464
+ }
+ ],
+ "md5sum": "f60d1b568010e488f1cb63f2d6be100a"
+ },
+ {
+ "dataPath": "params_shard_43.bin",
+ "format": "raw-shard",
+ "nbytes": 27712512,
+ "records": [
+ {
+ "name": "model.layers.33.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.34.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 829440
+ },
+ {
+ "name": "model.layers.34.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 834048
+ },
+ {
+ "name": "model.layers.34.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 8796672
+ },
+ {
+ "name": "model.layers.34.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 9792000
+ },
+ {
+ "name": "model.layers.34.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 12446208
+ },
+ {
+ "name": "model.layers.34.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 12777984
+ },
+ {
+ "name": "model.layers.34.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 12782592
+ },
+ {
+ "name": "model.layers.34.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 26053632
+ }
+ ],
+ "md5sum": "028cb8eb3d6eebee9e8aaaed2c85a8a5"
+ },
+ {
+ "dataPath": "params_shard_44.bin",
+ "format": "raw-shard",
+ "nbytes": 32689152,
+ "records": [
+ {
+ "name": "model.layers.34.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.34.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 6635520
+ },
+ {
+ "name": "model.layers.35.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 7464960
+ },
+ {
+ "name": "model.layers.35.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 7469568
+ },
+ {
+ "name": "model.layers.35.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 15432192
+ },
+ {
+ "name": "model.layers.35.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 16427520
+ },
+ {
+ "name": "model.layers.35.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 19081728
+ },
+ {
+ "name": "model.layers.35.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 19413504
+ },
+ {
+ "name": "model.layers.35.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 19418112
+ }
+ ],
+ "md5sum": "f692b0bd09b30dd1a32f65b3ec15d6af"
+ },
+ {
+ "dataPath": "params_shard_45.bin",
+ "format": "raw-shard",
+ "nbytes": 21076992,
+ "records": [
+ {
+ "name": "model.layers.35.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.35.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 1658880
+ },
+ {
+ "name": "model.layers.35.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 8294400
+ },
+ {
+ "name": "model.layers.36.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 9123840
+ },
+ {
+ "name": "model.layers.36.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 9128448
+ },
+ {
+ "name": "model.layers.36.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 17091072
+ },
+ {
+ "name": "model.layers.36.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 18086400
+ },
+ {
+ "name": "model.layers.36.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 20740608
+ },
+ {
+ "name": "model.layers.36.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 21072384
+ }
+ ],
+ "md5sum": "8bb3b9a458f2ddbada9d92e17c2ad4db"
+ },
+ {
+ "dataPath": "params_shard_46.bin",
+ "format": "raw-shard",
+ "nbytes": 31357440,
+ "records": [
+ {
+ "name": "model.layers.36.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.36.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 13271040
+ },
+ {
+ "name": "model.layers.36.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 14929920
+ },
+ {
+ "name": "model.layers.36.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 21565440
+ },
+ {
+ "name": "model.layers.37.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 22394880
+ },
+ {
+ "name": "model.layers.37.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 22399488
+ },
+ {
+ "name": "model.layers.37.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 30362112
+ }
+ ],
+ "md5sum": "05e52f40adae99c093fb9eeb4688fa08"
+ },
+ {
+ "dataPath": "params_shard_47.bin",
+ "format": "raw-shard",
+ "nbytes": 33352704,
+ "records": [
+ {
+ "name": "model.layers.37.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.37.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 2654208
+ },
+ {
+ "name": "model.layers.37.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 2985984
+ },
+ {
+ "name": "model.layers.37.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 2990592
+ },
+ {
+ "name": "model.layers.37.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 16261632
+ },
+ {
+ "name": "model.layers.37.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 17920512
+ },
+ {
+ "name": "model.layers.37.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 24556032
+ },
+ {
+ "name": "model.layers.38.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 25385472
+ },
+ {
+ "name": "model.layers.38.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 25390080
+ }
+ ],
+ "md5sum": "15ed1983becd2473774f24baef7afdc8"
+ },
+ {
+ "dataPath": "params_shard_48.bin",
+ "format": "raw-shard",
+ "nbytes": 26385408,
+ "records": [
+ {
+ "name": "model.layers.38.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.38.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 995328
+ },
+ {
+ "name": "model.layers.38.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 3649536
+ },
+ {
+ "name": "model.layers.38.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 3981312
+ },
+ {
+ "name": "model.layers.38.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 3985920
+ },
+ {
+ "name": "model.layers.38.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 17256960
+ },
+ {
+ "name": "model.layers.38.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 18915840
+ },
+ {
+ "name": "model.layers.38.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 25551360
+ },
+ {
+ "name": "model.layers.39.input_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 26380800
+ }
+ ],
+ "md5sum": "5b23b6d46d8ab4ffd4d295ca7d336f09"
+ },
+ {
+ "dataPath": "params_shard_49.bin",
+ "format": "raw-shard",
+ "nbytes": 33513984,
+ "records": [
+ {
+ "name": "model.layers.39.self_attn.wqkv_pack.q_weight",
+ "shape": [
+ 6912,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 7962624,
+ "byteOffset": 0
+ },
+ {
+ "name": "model.layers.39.self_attn.wqkv_pack.q_scale",
+ "shape": [
+ 6912,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 995328,
+ "byteOffset": 7962624
+ },
+ {
+ "name": "model.layers.39.self_attn.o_proj.q_weight",
+ "shape": [
+ 2304,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 2654208,
+ "byteOffset": 8957952
+ },
+ {
+ "name": "model.layers.39.self_attn.o_proj.q_scale",
+ "shape": [
+ 2304,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 331776,
+ "byteOffset": 11612160
+ },
+ {
+ "name": "model.layers.39.post_attention_layernorm.weight",
+ "shape": [
+ 2304
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 4608,
+ "byteOffset": 11943936
+ },
+ {
+ "name": "model.layers.39.mlp.gate_up_proj.q_weight",
+ "shape": [
+ 11520,
+ 288
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 13271040,
+ "byteOffset": 11948544
+ },
+ {
+ "name": "model.layers.39.mlp.gate_up_proj.q_scale",
+ "shape": [
+ 11520,
+ 72
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 1658880,
+ "byteOffset": 25219584
+ },
+ {
+ "name": "model.layers.39.mlp.down_proj.q_weight",
+ "shape": [
+ 2304,
+ 720
+ ],
+ "dtype": "uint32",
+ "format": "f32-to-bf16",
+ "nbytes": 6635520,
+ "byteOffset": 26878464
+ }
+ ],
+ "md5sum": "daa3f32720eb65cdc349f077aba714d7"
+ },
+ {
+ "dataPath": "params_shard_50.bin",
+ "format": "raw-shard",
+ "nbytes": 829440,
+ "records": [
+ {
+ "name": "model.layers.39.mlp.down_proj.q_scale",
+ "shape": [
+ 2304,
+ 180
+ ],
+ "dtype": "float16",
+ "format": "f32-to-bf16",
+ "nbytes": 829440,
+ "byteOffset": 0
+ }
+ ],
+ "md5sum": "6a7d2e49d33a040c1fe8533f541dd351"
+ }
+ ]
+}
\ No newline at end of file
diff --git a/params_shard_0.bin b/params_shard_0.bin
new file mode 100644
index 0000000000000000000000000000000000000000..a3872890f8733792f739153406d04f666f52750f
--- /dev/null
+++ b/params_shard_0.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:788c96a7fb1d616c48879ff2438e3f812a7fd42ae3ad6fcdf2af8efad1012fdb
+size 141419520
diff --git a/params_shard_1.bin b/params_shard_1.bin
new file mode 100644
index 0000000000000000000000000000000000000000..55fa4314dc18ac755e0bd4400efde676d3157bb8
--- /dev/null
+++ b/params_shard_1.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c412091b10fdc6e4f50f9637c66bed946b402b09de3ab89f4adff82a7a0636e8
+size 141419520
diff --git a/params_shard_10.bin b/params_shard_10.bin
new file mode 100644
index 0000000000000000000000000000000000000000..b86ccc5d297f6c4c7ad87336ea54508a100c7d0b
--- /dev/null
+++ b/params_shard_10.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46a0ebd07eff0a3b07c6e8e9e25df9429e4d1d4b8a7baa487602158652447fbd
+size 21076992
diff --git a/params_shard_11.bin b/params_shard_11.bin
new file mode 100644
index 0000000000000000000000000000000000000000..a595901ff6f7b448d2e89dab2c7c09f7bffb25c8
--- /dev/null
+++ b/params_shard_11.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b4e22689d4bd22dd5a1c6f584fdd8335c2cc2a322ca45cc853e033e4d8ac5291
+size 31357440
diff --git a/params_shard_12.bin b/params_shard_12.bin
new file mode 100644
index 0000000000000000000000000000000000000000..275ee45b55a344c9efbbe296cd668a306d692531
--- /dev/null
+++ b/params_shard_12.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:366eee63bacb9b0c34d0ecb00d425ad858d9e80203e2253184e6b1afbfdb5f59
+size 33352704
diff --git a/params_shard_13.bin b/params_shard_13.bin
new file mode 100644
index 0000000000000000000000000000000000000000..8582b5ad4e53f7c9f872e66d3434529b539df56e
--- /dev/null
+++ b/params_shard_13.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85791525b52d24e380faa29cc0dc1b90593586be68dc09fd91b3e94deea30ae2
+size 26385408
diff --git a/params_shard_14.bin b/params_shard_14.bin
new file mode 100644
index 0000000000000000000000000000000000000000..e0135fc5af67f0edfea5c5bb544b3631710f8ce6
--- /dev/null
+++ b/params_shard_14.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52ea8e415f9464edffd1ff362196202c7fd23f8695deb0fdfe2225c4e912e907
+size 33513984
diff --git a/params_shard_15.bin b/params_shard_15.bin
new file mode 100644
index 0000000000000000000000000000000000000000..ddb20b883f4ff6b716c019eee37e12468fb16d97
--- /dev/null
+++ b/params_shard_15.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f7c79c80a8a267c2e8f7ac1f213d33491f0e55b60b7bd802f9023517d9c87df
+size 27712512
diff --git a/params_shard_16.bin b/params_shard_16.bin
new file mode 100644
index 0000000000000000000000000000000000000000..9564c6ea94de822833c27cd0f4e74e2e9a2aeff6
--- /dev/null
+++ b/params_shard_16.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9134eb54dcf80c75f8c1d4836498c37c7f0248abee7b7e309dc7ffc14330e3c2
+size 32689152
diff --git a/params_shard_17.bin b/params_shard_17.bin
new file mode 100644
index 0000000000000000000000000000000000000000..b79a9bba7c5df30b10e0fceedaf281d25ef4ab1c
--- /dev/null
+++ b/params_shard_17.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d78e96934a9844a571fc85b72653c8d34d85e7487c35e6a2f8c7c8ba83667f05
+size 21076992
diff --git a/params_shard_18.bin b/params_shard_18.bin
new file mode 100644
index 0000000000000000000000000000000000000000..b7577ac7df92f61f629bf3f9f716b729ca9b06a0
--- /dev/null
+++ b/params_shard_18.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb765bb543129bf756a8889ce6b5023e57264f16986bd212b96efc7b2753136d
+size 31357440
diff --git a/params_shard_19.bin b/params_shard_19.bin
new file mode 100644
index 0000000000000000000000000000000000000000..1ca0cf54628ada2ab43b55cfd7e09004ab108d22
--- /dev/null
+++ b/params_shard_19.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc2a0179c74a01ce6dc9cb1766ecc652c5fd31ef87e89e88d0cf6f129a921d5e
+size 33352704
diff --git a/params_shard_2.bin b/params_shard_2.bin
new file mode 100644
index 0000000000000000000000000000000000000000..683f8b589bb71e0f8b48e8dcff638eae835bb7b0
--- /dev/null
+++ b/params_shard_2.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a19a09df01caae4c6ffee9bcb962b19c886db4a60a9687340e38478dd212154
+size 17677440
diff --git a/params_shard_20.bin b/params_shard_20.bin
new file mode 100644
index 0000000000000000000000000000000000000000..d92e07354b46b261f44815c54a961a3a994f17b9
--- /dev/null
+++ b/params_shard_20.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:484790c119fa646ec45fc6e01f772dea8bf4424ae708bd3f1e49cf9de52ce4e1
+size 26385408
diff --git a/params_shard_21.bin b/params_shard_21.bin
new file mode 100644
index 0000000000000000000000000000000000000000..25309876c680858ff97834b03d6c2427edcb8ef3
--- /dev/null
+++ b/params_shard_21.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb94c79676d81494a6ae98371f4c26352384e59aa9439170757b988e7981abcd
+size 33513984
diff --git a/params_shard_22.bin b/params_shard_22.bin
new file mode 100644
index 0000000000000000000000000000000000000000..cbd77430d7f70d18444a8e65ed960a6d1c762209
--- /dev/null
+++ b/params_shard_22.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db7b9396a43a8f780ee92344c3cfdcfb9398fb24f7b8c02cd8a0f0ffec94f16a
+size 27712512
diff --git a/params_shard_23.bin b/params_shard_23.bin
new file mode 100644
index 0000000000000000000000000000000000000000..9c9c1e5203b04310ce41dab25e7d9eaba5c7506f
--- /dev/null
+++ b/params_shard_23.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91db43613f2b95a8dbd628d97bbc9775656b58b50d107c0744fc2c9371ffb1b5
+size 32689152
diff --git a/params_shard_24.bin b/params_shard_24.bin
new file mode 100644
index 0000000000000000000000000000000000000000..a3d2324274a96b6e6bc1b7e7c9289f1f29cca07f
--- /dev/null
+++ b/params_shard_24.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de0e954bef47ebeac70ffbd5d6356682bfab6b0c6c58db0d1d9aa07777715671
+size 21076992
diff --git a/params_shard_25.bin b/params_shard_25.bin
new file mode 100644
index 0000000000000000000000000000000000000000..96b59a379c870c8d688fae0b4898d9e9fa38fe43
--- /dev/null
+++ b/params_shard_25.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2aef3c34529a2d45b87fe8f3808993be3f4005a362ddd79da74c0699999ccc11
+size 31357440
diff --git a/params_shard_26.bin b/params_shard_26.bin
new file mode 100644
index 0000000000000000000000000000000000000000..1243db10aaf6a081ebe83bab1c54183284d0747c
--- /dev/null
+++ b/params_shard_26.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b97525ad67383b0459e2ba2bcb729314fdb23db6a2ac1cb4c17e1ed0a373512
+size 33352704
diff --git a/params_shard_27.bin b/params_shard_27.bin
new file mode 100644
index 0000000000000000000000000000000000000000..3cec869746bd5c645befc6f944b72ecb2ddf86a0
--- /dev/null
+++ b/params_shard_27.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8475a7c5db303992116230541e84947e4bc2786a29f3fa6ad021bfccfe95fe08
+size 26385408
diff --git a/params_shard_28.bin b/params_shard_28.bin
new file mode 100644
index 0000000000000000000000000000000000000000..e138e374952e45d103dd1a875608c8572267bf44
--- /dev/null
+++ b/params_shard_28.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8dddeb4b0d4f2a8bbcb0ce39f857e96be70456ca0e6e74962bd722432c3fff0f
+size 33513984
diff --git a/params_shard_29.bin b/params_shard_29.bin
new file mode 100644
index 0000000000000000000000000000000000000000..dc06419e257f85c6c14574d18b65d62d76798675
--- /dev/null
+++ b/params_shard_29.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf5ee6e4298c640c7ec70bd5a23d11f04c51d02fdfb7148c4501ff937920a49c
+size 27712512
diff --git a/params_shard_3.bin b/params_shard_3.bin
new file mode 100644
index 0000000000000000000000000000000000000000..4fba861388d717793d1ae519c2bd95fbe79b1eee
--- /dev/null
+++ b/params_shard_3.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:78c6565133a92a1f6db275e5bf9ad875bb2632f8ab13093dd6fd6be88bbdf74c
+size 29635200
diff --git a/params_shard_30.bin b/params_shard_30.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f529bc64d44642fe74f0122fdafde455dac3ba2b
--- /dev/null
+++ b/params_shard_30.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:479ccf1eb31f8bfbacb261b2d76bd61fad81a44d696156029960b14e85325ef5
+size 32689152
diff --git a/params_shard_31.bin b/params_shard_31.bin
new file mode 100644
index 0000000000000000000000000000000000000000..a646d9cb70a2224331de6eed75c6c0b11110874f
--- /dev/null
+++ b/params_shard_31.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7f6d6571a4d3b58767c49f36b0c4192e6a81e2a4e68a5c6c060bd5e7540e192
+size 21076992
diff --git a/params_shard_32.bin b/params_shard_32.bin
new file mode 100644
index 0000000000000000000000000000000000000000..75b5e1792245e3820c2f56cd31c75fbe1adfd05c
--- /dev/null
+++ b/params_shard_32.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa2b0c8fc0527d9d9efdc5ad82a455198c8e3b4106b176c5997c297857af83cc
+size 31357440
diff --git a/params_shard_33.bin b/params_shard_33.bin
new file mode 100644
index 0000000000000000000000000000000000000000..e893055a29d6cd1ac60a2ade704fca6d010deeab
--- /dev/null
+++ b/params_shard_33.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eac50a0274dde0a951c397117a8be614f51719f94f43f4e581b923d4f2bd7cbf
+size 33352704
diff --git a/params_shard_34.bin b/params_shard_34.bin
new file mode 100644
index 0000000000000000000000000000000000000000..7c05cc2700f9b302116e5a3f1e664c89342cb11e
--- /dev/null
+++ b/params_shard_34.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e05e7ce6f87b2068c03c9eead0f9ead3939934cfc7d9001fc9ab2aca46586491
+size 26385408
diff --git a/params_shard_35.bin b/params_shard_35.bin
new file mode 100644
index 0000000000000000000000000000000000000000..1be3046c37047066a1d904b083f3cfaa76144151
--- /dev/null
+++ b/params_shard_35.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d15490a48a9b92c94bd14f267ae4454f5e4da9980907247124a243f3c5bbb2a
+size 33513984
diff --git a/params_shard_36.bin b/params_shard_36.bin
new file mode 100644
index 0000000000000000000000000000000000000000..155ff739e12da855066cb464ea91a5ddc3ec49d9
--- /dev/null
+++ b/params_shard_36.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ee4ed3bdfa97486bc5d8b73f81268887fa5e5e8bdea7d024a1354c594ef37ee
+size 27712512
diff --git a/params_shard_37.bin b/params_shard_37.bin
new file mode 100644
index 0000000000000000000000000000000000000000..2287da32b8df498fc5d4bf5bdb399336e14fcb6d
--- /dev/null
+++ b/params_shard_37.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53f19f44caa9a4844d892605df8b2719b11e53443f87dcaa4d54c20b66cd713a
+size 32689152
diff --git a/params_shard_38.bin b/params_shard_38.bin
new file mode 100644
index 0000000000000000000000000000000000000000..c64488b258ec3c8dcc533c562ba3a36a889828e6
--- /dev/null
+++ b/params_shard_38.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a8862c14ca2ebfce485314db90f5a8e7f4d7eac0f2f5d8a218ee4f0f13999fc
+size 21076992
diff --git a/params_shard_39.bin b/params_shard_39.bin
new file mode 100644
index 0000000000000000000000000000000000000000..37984e074b0aa82e88df56b7dc04b18f1777239e
--- /dev/null
+++ b/params_shard_39.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d4f915bad9724bae0bc018b185b1a1ca3cd62c9fedb9ffe0dad3a00e27ceadf
+size 31357440
diff --git a/params_shard_4.bin b/params_shard_4.bin
new file mode 100644
index 0000000000000000000000000000000000000000..2ef0c885c7c010b00b8fa12ae3995b0c44186af5
--- /dev/null
+++ b/params_shard_4.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:67a89a96db2e2e49ab0aa603d3b2475ab6f17a972dd283cde23e32cc0f4c68fe
+size 31357440
diff --git a/params_shard_40.bin b/params_shard_40.bin
new file mode 100644
index 0000000000000000000000000000000000000000..d1dea345987323b05b02a9a9f06a50bda36606ef
--- /dev/null
+++ b/params_shard_40.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3425f9b93fdfbcfef0c2f6eddec37f2fce23b5dac904ea750a9ea06cde60c9e0
+size 33352704
diff --git a/params_shard_41.bin b/params_shard_41.bin
new file mode 100644
index 0000000000000000000000000000000000000000..1319bd810634fcef845b76f38b1bf3fcb3aa586d
--- /dev/null
+++ b/params_shard_41.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:56f8a31dc967b3e18bc361205a1799b9c003b701ce29307b934fbfb9e9251df8
+size 26385408
diff --git a/params_shard_42.bin b/params_shard_42.bin
new file mode 100644
index 0000000000000000000000000000000000000000..db545293d6fde7453d06cf29a7e1a4d80e0c7812
--- /dev/null
+++ b/params_shard_42.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da3b0aeb0d7cf27d4cdaa17781a4ba67d0be0ec3c07d17d1ca320f471cb7f720
+size 33513984
diff --git a/params_shard_43.bin b/params_shard_43.bin
new file mode 100644
index 0000000000000000000000000000000000000000..cca163e480a443f6d3469136777d429d1123ccce
--- /dev/null
+++ b/params_shard_43.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bbf0ae509288a346778f1c5e928bbd686fa130c900212e51533792ab4b4acbbb
+size 27712512
diff --git a/params_shard_44.bin b/params_shard_44.bin
new file mode 100644
index 0000000000000000000000000000000000000000..9d9a0385a1763609295c21cc3f81529fd0820f26
--- /dev/null
+++ b/params_shard_44.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db3f3f146ea586e44aa23afbbfdf9998da732ee6c11c494af28c8846ec48a38e
+size 32689152
diff --git a/params_shard_45.bin b/params_shard_45.bin
new file mode 100644
index 0000000000000000000000000000000000000000..b66ec372258f5104d3b43cbbf885446f04b70425
--- /dev/null
+++ b/params_shard_45.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6eafb0d112e9ba2d2994a34ebd777c5ff1e5f20da30bca07a4a8adb4585f6f3a
+size 21076992
diff --git a/params_shard_46.bin b/params_shard_46.bin
new file mode 100644
index 0000000000000000000000000000000000000000..e54dd4253474ff158c69369f48a1d5681b7772cf
--- /dev/null
+++ b/params_shard_46.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5304371c2f62ab921ed21907fc6df9fd62732450a5063ddc4705c1dbbc845530
+size 31357440
diff --git a/params_shard_47.bin b/params_shard_47.bin
new file mode 100644
index 0000000000000000000000000000000000000000..1e9ad82ca1c4540335e0645a9aae073ebfcaa0f0
--- /dev/null
+++ b/params_shard_47.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a0c9f3a8a59683961a43bafa51d8848004b9b493a4cec41b01baed979850f85f
+size 33352704
diff --git a/params_shard_48.bin b/params_shard_48.bin
new file mode 100644
index 0000000000000000000000000000000000000000..12c8078bdc0bdb53ce6d34226b78835560e1ccf6
--- /dev/null
+++ b/params_shard_48.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:08159db3709da583e38cfc45dc64369c645e5c1cf61ca192ad08e47efc8e3a06
+size 26385408
diff --git a/params_shard_49.bin b/params_shard_49.bin
new file mode 100644
index 0000000000000000000000000000000000000000..b43d8270640c2f2aaede9bff2acefeaa5a7240ec
--- /dev/null
+++ b/params_shard_49.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9de76b3602c4c55aa15c726e4b1609426a8d14e8ec5994b0b6c170d679787e6e
+size 33513984
diff --git a/params_shard_5.bin b/params_shard_5.bin
new file mode 100644
index 0000000000000000000000000000000000000000..3345674c5cc1ee836c102a9346fc9db934e511ac
--- /dev/null
+++ b/params_shard_5.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8c0ad0acdd78f8a49d9e92b6799f4a14cda2d1cbbc8043de6f71bb2d48421f23
+size 33352704
diff --git a/params_shard_50.bin b/params_shard_50.bin
new file mode 100644
index 0000000000000000000000000000000000000000..042a0a622e3d06069edb1641306626b7a6729496
--- /dev/null
+++ b/params_shard_50.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b85d6d8e3c959fd554614d649c3b0557d203dd23b5e938530b9c2ff125d5d0d
+size 829440
diff --git a/params_shard_6.bin b/params_shard_6.bin
new file mode 100644
index 0000000000000000000000000000000000000000..264ac84a67613ea8da31226230dcf4df14f79bbe
--- /dev/null
+++ b/params_shard_6.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c52edf877d7e19aff18d34b58920922d162a072551bcb9a538aad4147734754
+size 26385408
diff --git a/params_shard_7.bin b/params_shard_7.bin
new file mode 100644
index 0000000000000000000000000000000000000000..b8658c2bf80f69404055d9f2a523233ea1123047
--- /dev/null
+++ b/params_shard_7.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91e4c0fddd1d86e9e0a7cdad9220f9315f922172aa0df8cec0a938a76ac71d16
+size 33513984
diff --git a/params_shard_8.bin b/params_shard_8.bin
new file mode 100644
index 0000000000000000000000000000000000000000..1a20ff9552e8e90f1f1da20b31290c05146538b1
--- /dev/null
+++ b/params_shard_8.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:32ddaed1311aa643624f6a1f7891f4a319100956d93cc71dbb83a903b96be5fd
+size 27712512
diff --git a/params_shard_9.bin b/params_shard_9.bin
new file mode 100644
index 0000000000000000000000000000000000000000..b8ebf9b773aaa8a93c7f35f92316e2991f5bf56f
--- /dev/null
+++ b/params_shard_9.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ddc524dbbc578b9a523431aefb85b750d107753d80492dfe03fa8db6b4b59c5e
+size 32689152
diff --git a/tokenizer.json b/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..21b7d2808648961dd8cf490cd6cca0567142248d
--- /dev/null
+++ b/tokenizer.json
@@ -0,0 +1,294498 @@
+{
+ "version": "1.0",
+ "truncation": null,
+ "padding": null,
+ "added_tokens": [
+ {
+ "id": 0,
+ "content": "",
+ "single_word": false,
+ "lstrip": false,
+ "rstrip": false,
+ "normalized": false,
+ "special": true
+ },
+ {
+ "id": 1,
+ "content": "",
+ "single_word": false,
+ "lstrip": false,
+ "rstrip": false,
+ "normalized": false,
+ "special": true
+ },
+ {
+ "id": 2,
+ "content": "",
+ "single_word": false,
+ "lstrip": false,
+ "rstrip": false,
+ "normalized": false,
+ "special": true
+ },
+ {
+ "id": 122753,
+ "content": "<|im_end|>",
+ "single_word": false,
+ "lstrip": false,
+ "rstrip": false,
+ "normalized": false,
+ "special": true
+ },
+ {
+ "id": 122754,
+ "content": "▁",
+ "single_word": false,
+ "lstrip": false,
+ "rstrip": false,
+ "normalized": false,
+ "special": true
+ },
+ {
+ "id": 122755,
+ "content": "▁",
+ "single_word": false,
+ "lstrip": false,
+ "rstrip": false,
+ "normalized": false,
+ "special": true
+ },
+ {
+ "id": 122756,
+ "content": "<|tool_call|>",
+ "single_word": false,
+ "lstrip": false,
+ "rstrip": false,
+ "normalized": false,
+ "special": true
+ },
+ {
+ "id": 122757,
+ "content": "<|im_start|>",
+ "single_word": false,
+ "lstrip": false,
+ "rstrip": false,
+ "normalized": false,
+ "special": true
+ },
+ {
+ "id": 122758,
+ "content": "▁",
+ "single_word": false,
+ "lstrip": false,
+ "rstrip": false,
+ "normalized": false,
+ "special": true
+ },
+ {
+ "id": 122759,
+ "content": "▁",
+ "single_word": false,
+ "lstrip": false,
+ "rstrip": false,
+ "normalized": false,
+ "special": true
+ }
+ ],
+ "normalizer": {
+ "type": "Sequence",
+ "normalizers": [
+ {
+ "type": "Prepend",
+ "prepend": "▁"
+ },
+ {
+ "type": "Replace",
+ "pattern": {
+ "String": " "
+ },
+ "content": "▁"
+ }
+ ]
+ },
+ "pre_tokenizer": null,
+ "post_processor": {
+ "type": "TemplateProcessing",
+ "single": [
+ {
+ "SpecialToken": {
+ "id": "",
+ "type_id": 0
+ }
+ },
+ {
+ "Sequence": {
+ "id": "A",
+ "type_id": 0
+ }
+ }
+ ],
+ "pair": [
+ {
+ "SpecialToken": {
+ "id": "",
+ "type_id": 0
+ }
+ },
+ {
+ "Sequence": {
+ "id": "A",
+ "type_id": 0
+ }
+ },
+ {
+ "SpecialToken": {
+ "id": "",
+ "type_id": 1
+ }
+ },
+ {
+ "Sequence": {
+ "id": "B",
+ "type_id": 1
+ }
+ }
+ ],
+ "special_tokens": {
+ "": {
+ "id": "",
+ "ids": [
+ 1
+ ],
+ "tokens": [
+ ""
+ ]
+ }
+ }
+ },
+ "decoder": {
+ "type": "Sequence",
+ "decoders": [
+ {
+ "type": "Replace",
+ "pattern": {
+ "String": "▁"
+ },
+ "content": " "
+ },
+ {
+ "type": "ByteFallback"
+ },
+ {
+ "type": "Fuse"
+ },
+ {
+ "type": "Strip",
+ "content": " ",
+ "start": 1,
+ "stop": 0
+ }
+ ]
+ },
+ "model": {
+ "type": "BPE",
+ "dropout": null,
+ "unk_token": "",
+ "continuing_subword_prefix": null,
+ "end_of_word_suffix": null,
+ "fuse_unk": true,
+ "byte_fallback": true,
+ "vocab": {
+ "": 0,
+ "": 1,
+ "": 2,
+ "": 3,
+ "": 4,
+ "\n": 5,
+ "\t": 6,
+ "
": 7,
+ "
": 8,
+ "": 9,
+ "": 10,
+ "": 11,
+ "
": 12,
+ "": 13,
+ " | | ": 14,
+ "": 15,
+ "": 16,
+ "": 17,
+ "": 18,
+ "": 21,
+ "": 22,
+ "
": 23,
+ "": 24,
+ "": 25,
+ "": 26,
+ "": 27,
+ "": 28,
+ "": 29,
+ "": 30,
+ "": 31,
+ "": 32,
+ "
": 33,
+ "
": 34,
+ "
": 35,
+ "": 36,
+ "": 37,
+ "": 38,
+ "
": 39,
+ "": 40,
+ "": 41,
+ "
": 42,
+ "": 43,
+ "
": 44,
+ "
": 45,
+ "": 46,
+ "": 47,
+ "
": 48,
+ "": 49,
+ "": 50,
+ "": 51,
+ "0": 52,
+ "1": 53,
+ "2": 54,
+ "3": 55,
+ "4": 56,
+ "5": 57,
+ "6": 58,
+ "7": 59,
+ "8": 60,
+ "9": 61,
+ "+": 62,
+ "-": 63,
+ "=": 64,
+ ",": 65,
+ "。": 66,
+ "!": 67,
+ "?": 68,
+ "、": 69,
+ ":": 70,
+ "¥": 71,
+ ".": 72,
+ "!": 73,
+ "?": 74,
+ "...": 75,
+ "。。。": 76,
+ "。。。。。。": 77,
+ "《": 78,
+ "》": 79,
+ "【": 80,
+ "】": 81,
+ "『": 82,
+ "』": 83,
+ "```": 84,
+ "": 86,
+ "---": 87,
+ "": 88,
+ ";": 89,
+ ".": 90,
+ "=": 91,
+ "<": 92,
+ ">": 93,
+ "-": 94,
+ "+": 95,
+ "%": 96,
+ "‼": 97,
+ "㊣": 98,
+ "/": 99,
+ "|": 100,
+ "": 101,
+ "": 102,
+ "": 103,
+ "": 104,
+ "": 105,
+ "": 106,
+ "": 107,
+ "": 108,
+ "": 109,
+ "": 110,
+ "": 111,
+ "": 112,
+ "": 113,
+ "": 114,
+ "": 115,
+ "": 116,
+ "": 117,
+ "": 118,
+ "": 119,
+ "": 120,
+ "": 121,
+ "": 122,
+ "": 123,
+ "": 124,
+ "": 125,
+ "": 126,
+ "": 127,
+ "": 128,
+ "": 129,
+ "": 130,
+ "": 131,
+ "": 132,
+ "": 133,
+ "": 134,
+ "": 135,
+ "": 136,
+ "": 137,
+ "": 138,
+ "": 139,
+ "": 140,
+ "": 141,
+ "": 142,
+ "": 143,
+ "": 144,
+ "": 145,
+ "": 146,
+ "": 147,
+ "": 148,
+ "": 149,
+ "": 150,
+ "": 151,
+ "": 152,
+ "": 153,
+ "": 154,
+ "": 155,
+ "": 156,
+ "": 157,
+ "": 158,
+ "": 159,
+ "": 160,
+ "": 161,
+ "": 162,
+ "": 163,
+ "": 164,
+ "": 165,
+ "": 166,
+ "": 167,
+ "": 168,
+ "": 169,
+ "": 170,
+ "": 171,
+ "": 172,
+ "": 173,
+ "": 174,
+ "": 175,
+ "": 176,
+ "": 177,
+ "": 178,
+ "": 179,
+ "": 180,
+ "": 181,
+ "": 182,
+ "": 183,
+ "": 184,
+ "": 185,
+ "": 186,
+ "": 187,
+ "": 188,
+ "": 189,
+ "": 190,
+ "": 191,
+ "": 192,
+ "": 193,
+ "": 194,
+ "": 195,
+ "": 196,
+ "": 197,
+ "": 198,
+ "": 199,
+ "": 200,
+ "": 201,
+ "": 202,
+ "": 203,
+ "": 204,
+ "": 205,
+ "": 206,
+ "": 207,
+ "": 208,
+ "": 209,
+ "": 210,
+ "": 211,
+ "": 212,
+ "": 213,
+ "": 214,
+ "": 215,
+ "": 216,
+ "": 217,
+ "": 218,
+ "": 219,
+ "": 220,
+ "": 221,
+ "": 222,
+ "": 223,
+ "": 224,
+ "": 225,
+ "": 226,
+ "": 227,
+ "": 228,
+ "": 229,
+ "": 230,
+ "": 231,
+ "": 232,
+ "": 233,
+ "": 234,
+ "": 235,
+ "": 236,
+ "": 237,
+ "": 238,
+ "": 239,
+ "": 240,
+ "": 241,
+ "": 242,
+ "": 243,
+ "": 244,
+ "": 245,
+ "": 246,
+ "": 247,
+ "": 248,
+ "": 249,
+ "": 250,
+ "": 251,
+ "": 252,
+ "": 253,
+ "": 254,
+ "": 255,
+ "": 256,
+ "": 257,
+ "": 258,
+ "": 259,
+ "": 260,
+ "": 261,
+ "": 262,
+ "": 263,
+ "": 264,
+ "": 265,
+ "": 266,
+ "": 267,
+ "": 268,
+ "": 269,
+ "": 270,
+ "": 271,
+ "": 272,
+ "": 273,
+ "": 274,
+ "": 275,
+ "": 276,
+ "": 277,
+ "": 278,
+ "": 279,
+ "": 280,
+ "": 281,
+ "": 282,
+ "": 283,
+ "": 284,
+ "": 285,
+ "": 286,
+ "": 287,
+ "": 288,
+ "": 289,
+ "": 290,
+ "": 291,
+ "": 292,
+ "": 293,
+ "": 294,
+ "": 295,
+ "": 296,
+ "": 297,
+ "": 298,
+ "": 299,
+ "": 300,
+ "": 301,
+ "": 302,
+ "": 303,
+ "": 304,
+ "": 305,
+ "": 306,
+ "": 307,
+ "": 308,
+ "": 309,
+ "": 310,
+ "": 311,
+ "": 312,
+ "": 313,
+ "": 314,
+ "": 315,
+ "": 316,
+ "": 317,
+ "": 318,
+ "": 319,
+ "": 320,
+ "": 321,
+ "": 322,
+ "": 323,
+ "": 324,
+ "": 325,
+ "": 326,
+ "": 327,
+ "": 328,
+ "": 329,
+ "": 330,
+ "": 331,
+ "": 332,
+ "": 333,
+ "": 334,
+ "": 335,
+ "": 336,
+ "": 337,
+ "": 338,
+ "": 339,
+ "": 340,
+ "": 341,
+ "": 342,
+ "": 343,
+ "": 344,
+ "": 345,
+ "": 346,
+ "": 347,
+ "": 348,
+ "": 349,
+ "": 350,
+ "": 351,
+ "": 352,
+ "": 353,
+ "