Hugging Face's logo

mlc-ai
/
Llama-3.2-3B-Instruct-q0f32-MLC

MLC-LLM
web-llm
Model card Files Community
CharlieFRuan commited on
Commit
1363d6b
•
1 Parent(s): f54309d

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. mlc-chat-config.json +90 -0
  2. ndarray-cache-b16.json +2726 -0
  3. ndarray-cache.json +2726 -0
  4. params_shard_0.bin +3 -0
  5. params_shard_1.bin +3 -0
  6. params_shard_10.bin +3 -0
  7. params_shard_100.bin +3 -0
  8. params_shard_101.bin +3 -0
  9. params_shard_102.bin +3 -0
  10. params_shard_103.bin +3 -0
  11. params_shard_104.bin +3 -0
  12. params_shard_105.bin +3 -0
  13. params_shard_106.bin +3 -0
  14. params_shard_107.bin +3 -0
  15. params_shard_108.bin +3 -0
  16. params_shard_109.bin +3 -0
  17. params_shard_11.bin +3 -0
  18. params_shard_110.bin +3 -0
  19. params_shard_111.bin +3 -0
  20. params_shard_112.bin +3 -0
  21. params_shard_12.bin +3 -0
  22. params_shard_13.bin +3 -0
  23. params_shard_14.bin +3 -0
  24. params_shard_15.bin +3 -0
  25. params_shard_16.bin +3 -0
  26. params_shard_17.bin +3 -0
  27. params_shard_18.bin +3 -0
  28. params_shard_19.bin +3 -0
  29. params_shard_2.bin +3 -0
  30. params_shard_20.bin +3 -0
  31. params_shard_21.bin +3 -0
  32. params_shard_22.bin +3 -0
  33. params_shard_23.bin +3 -0
  34. params_shard_24.bin +3 -0
  35. params_shard_25.bin +3 -0
  36. params_shard_26.bin +3 -0
  37. params_shard_27.bin +3 -0
  38. params_shard_28.bin +3 -0
  39. params_shard_29.bin +3 -0
  40. params_shard_3.bin +3 -0
  41. params_shard_30.bin +3 -0
  42. params_shard_31.bin +3 -0
  43. params_shard_32.bin +3 -0
  44. params_shard_33.bin +3 -0
  45. params_shard_34.bin +3 -0
  46. params_shard_35.bin +3 -0
  47. params_shard_36.bin +3 -0
  48. params_shard_37.bin +3 -0
  49. params_shard_38.bin +3 -0
  50. params_shard_39.bin +3 -0
mlc-chat-config.json ADDED
@@ -0,0 +1,90 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "version": "0.1.0",
3
+ "model_type": "llama",
4
+ "quantization": "q0f32",
5
+ "model_config": {
6
+ "hidden_size": 3072,
7
+ "intermediate_size": 8192,
8
+ "num_attention_heads": 24,
9
+ "num_hidden_layers": 28,
10
+ "rms_norm_eps": 1e-05,
11
+ "vocab_size": 128256,
12
+ "tie_word_embeddings": true,
13
+ "position_embedding_base": 500000.0,
14
+ "rope_scaling": {
15
+ "factor": 32.0,
16
+ "high_freq_factor": 4.0,
17
+ "low_freq_factor": 1.0,
18
+ "original_max_position_embeddings": 8192,
19
+ "rope_type": "llama3"
20
+ },
21
+ "context_window_size": 131072,
22
+ "prefill_chunk_size": 8192,
23
+ "num_key_value_heads": 8,
24
+ "head_dim": 128,
25
+ "tensor_parallel_shards": 1,
26
+ "pipeline_parallel_stages": 1,
27
+ "max_batch_size": 128
28
+ },
29
+ "vocab_size": 128256,
30
+ "context_window_size": 131072,
31
+ "sliding_window_size": -1,
32
+ "prefill_chunk_size": 8192,
33
+ "attention_sink_size": -1,
34
+ "tensor_parallel_shards": 1,
35
+ "pipeline_parallel_stages": 1,
36
+ "temperature": 0.6,
37
+ "presence_penalty": 0.0,
38
+ "frequency_penalty": 0.0,
39
+ "repetition_penalty": 1.0,
40
+ "top_p": 0.9,
41
+ "tokenizer_files": [
42
+ "tokenizer.json",
43
+ "tokenizer_config.json"
44
+ ],
45
+ "tokenizer_info": {
46
+ "token_postproc_method": "byte_level",
47
+ "prepend_space_in_encode": false,
48
+ "strip_space_in_decode": false
49
+ },
50
+ "conv_template": {
51
+ "name": "llama-3_1",
52
+ "system_template": "<|start_header_id|>system<|end_header_id|>\n\n{system_message}<|eot_id|>",
53
+ "system_message": "You are a helpful, respectful and honest assistant.",
54
+ "system_prefix_token_ids": [
55
+ 128000
56
+ ],
57
+ "add_role_after_system_message": true,
58
+ "roles": {
59
+ "user": "<|start_header_id|>user",
60
+ "assistant": "<|start_header_id|>assistant",
61
+ "tool": "<|start_header_id|>ipython"
62
+ },
63
+ "role_templates": {
64
+ "user": "{user_message}",
65
+ "assistant": "{assistant_message}",
66
+ "tool": "{tool_message}"
67
+ },
68
+ "messages": [],
69
+ "seps": [
70
+ "<|eot_id|>"
71
+ ],
72
+ "role_content_sep": "<|end_header_id|>\n\n",
73
+ "role_empty_sep": "<|end_header_id|>\n\n",
74
+ "stop_str": [],
75
+ "stop_token_ids": [
76
+ 128001,
77
+ 128008,
78
+ 128009
79
+ ],
80
+ "function_string": "",
81
+ "use_function_calling": false
82
+ },
83
+ "pad_token_id": 0,
84
+ "bos_token_id": 128000,
85
+ "eos_token_id": [
86
+ 128001,
87
+ 128008,
88
+ 128009
89
+ ]
90
+ }
ndarray-cache-b16.json ADDED
@@ -0,0 +1,2726 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "ParamSize": 170,
4
+ "ParamBytes": 12850999296.0,
5
+ "BitsPerParam": 32.0
6
+ },
7
+ "records": [
8
+ {
9
+ "dataPath": "params_shard_0.bin",
10
+ "format": "raw-shard",
11
+ "nbytes": 788004864,
12
+ "records": [
13
+ {
14
+ "name": "model.embed_tokens.weight",
15
+ "shape": [
16
+ 128256,
17
+ 3072
18
+ ],
19
+ "dtype": "bfloat16",
20
+ "format": "raw",
21
+ "nbytes": 788004864,
22
+ "byteOffset": 0
23
+ }
24
+ ],
25
+ "md5sum": "133c0866b77e2e5678be91a753bb7e28"
26
+ },
27
+ {
28
+ "dataPath": "params_shard_1.bin",
29
+ "format": "raw-shard",
30
+ "nbytes": 50331648,
31
+ "records": [
32
+ {
33
+ "name": "model.layers.0.mlp.down_proj.weight",
34
+ "shape": [
35
+ 3072,
36
+ 8192
37
+ ],
38
+ "dtype": "bfloat16",
39
+ "format": "raw",
40
+ "nbytes": 50331648,
41
+ "byteOffset": 0
42
+ }
43
+ ],
44
+ "md5sum": "93080db9b85b8252e1138672b9902f9a"
45
+ },
46
+ {
47
+ "dataPath": "params_shard_2.bin",
48
+ "format": "raw-shard",
49
+ "nbytes": 100663296,
50
+ "records": [
51
+ {
52
+ "name": "model.layers.0.mlp.gate_up_proj.weight",
53
+ "shape": [
54
+ 16384,
55
+ 3072
56
+ ],
57
+ "dtype": "bfloat16",
58
+ "format": "raw",
59
+ "nbytes": 100663296,
60
+ "byteOffset": 0
61
+ }
62
+ ],
63
+ "md5sum": "0c18f71e770e7c7c07270f2d7ec06d1c"
64
+ },
65
+ {
66
+ "dataPath": "params_shard_3.bin",
67
+ "format": "raw-shard",
68
+ "nbytes": 18874368,
69
+ "records": [
70
+ {
71
+ "name": "model.layers.0.self_attn.o_proj.weight",
72
+ "shape": [
73
+ 3072,
74
+ 3072
75
+ ],
76
+ "dtype": "bfloat16",
77
+ "format": "raw",
78
+ "nbytes": 18874368,
79
+ "byteOffset": 0
80
+ }
81
+ ],
82
+ "md5sum": "44251b37a994062bedadfcd599328de2"
83
+ },
84
+ {
85
+ "dataPath": "params_shard_4.bin",
86
+ "format": "raw-shard",
87
+ "nbytes": 50331648,
88
+ "records": [
89
+ {
90
+ "name": "model.layers.1.mlp.down_proj.weight",
91
+ "shape": [
92
+ 3072,
93
+ 8192
94
+ ],
95
+ "dtype": "bfloat16",
96
+ "format": "raw",
97
+ "nbytes": 50331648,
98
+ "byteOffset": 0
99
+ }
100
+ ],
101
+ "md5sum": "bf00d477f6d478010a9c7584c93723fd"
102
+ },
103
+ {
104
+ "dataPath": "params_shard_5.bin",
105
+ "format": "raw-shard",
106
+ "nbytes": 100663296,
107
+ "records": [
108
+ {
109
+ "name": "model.layers.1.mlp.gate_up_proj.weight",
110
+ "shape": [
111
+ 16384,
112
+ 3072
113
+ ],
114
+ "dtype": "bfloat16",
115
+ "format": "raw",
116
+ "nbytes": 100663296,
117
+ "byteOffset": 0
118
+ }
119
+ ],
120
+ "md5sum": "23a00478c5789710e44b2229f5b1f0ea"
121
+ },
122
+ {
123
+ "dataPath": "params_shard_6.bin",
124
+ "format": "raw-shard",
125
+ "nbytes": 31457280,
126
+ "records": [
127
+ {
128
+ "name": "model.layers.1.self_attn.qkv_proj.weight",
129
+ "shape": [
130
+ 5120,
131
+ 3072
132
+ ],
133
+ "dtype": "bfloat16",
134
+ "format": "raw",
135
+ "nbytes": 31457280,
136
+ "byteOffset": 0
137
+ }
138
+ ],
139
+ "md5sum": "c15617b1ce4de0ecc8e85d8566d4fd07"
140
+ },
141
+ {
142
+ "dataPath": "params_shard_7.bin",
143
+ "format": "raw-shard",
144
+ "nbytes": 18874368,
145
+ "records": [
146
+ {
147
+ "name": "model.layers.1.self_attn.o_proj.weight",
148
+ "shape": [
149
+ 3072,
150
+ 3072
151
+ ],
152
+ "dtype": "bfloat16",
153
+ "format": "raw",
154
+ "nbytes": 18874368,
155
+ "byteOffset": 0
156
+ }
157
+ ],
158
+ "md5sum": "7926239d43547e6b686b15951a3a84e7"
159
+ },
160
+ {
161
+ "dataPath": "params_shard_8.bin",
162
+ "format": "raw-shard",
163
+ "nbytes": 50331648,
164
+ "records": [
165
+ {
166
+ "name": "model.layers.10.mlp.down_proj.weight",
167
+ "shape": [
168
+ 3072,
169
+ 8192
170
+ ],
171
+ "dtype": "bfloat16",
172
+ "format": "raw",
173
+ "nbytes": 50331648,
174
+ "byteOffset": 0
175
+ }
176
+ ],
177
+ "md5sum": "2237119d8791ce2a58941361b80fa84d"
178
+ },
179
+ {
180
+ "dataPath": "params_shard_9.bin",
181
+ "format": "raw-shard",
182
+ "nbytes": 100663296,
183
+ "records": [
184
+ {
185
+ "name": "model.layers.10.mlp.gate_up_proj.weight",
186
+ "shape": [
187
+ 16384,
188
+ 3072
189
+ ],
190
+ "dtype": "bfloat16",
191
+ "format": "raw",
192
+ "nbytes": 100663296,
193
+ "byteOffset": 0
194
+ }
195
+ ],
196
+ "md5sum": "d70ec745e1feafbb4a1d86914adfca38"
197
+ },
198
+ {
199
+ "dataPath": "params_shard_10.bin",
200
+ "format": "raw-shard",
201
+ "nbytes": 31457280,
202
+ "records": [
203
+ {
204
+ "name": "model.layers.10.self_attn.qkv_proj.weight",
205
+ "shape": [
206
+ 5120,
207
+ 3072
208
+ ],
209
+ "dtype": "bfloat16",
210
+ "format": "raw",
211
+ "nbytes": 31457280,
212
+ "byteOffset": 0
213
+ }
214
+ ],
215
+ "md5sum": "75f3e0d41780aa8602f46b1016314c05"
216
+ },
217
+ {
218
+ "dataPath": "params_shard_11.bin",
219
+ "format": "raw-shard",
220
+ "nbytes": 18874368,
221
+ "records": [
222
+ {
223
+ "name": "model.layers.10.self_attn.o_proj.weight",
224
+ "shape": [
225
+ 3072,
226
+ 3072
227
+ ],
228
+ "dtype": "bfloat16",
229
+ "format": "raw",
230
+ "nbytes": 18874368,
231
+ "byteOffset": 0
232
+ }
233
+ ],
234
+ "md5sum": "90f3ce08908231682aa92f2c1296e5fc"
235
+ },
236
+ {
237
+ "dataPath": "params_shard_12.bin",
238
+ "format": "raw-shard",
239
+ "nbytes": 50331648,
240
+ "records": [
241
+ {
242
+ "name": "model.layers.11.mlp.down_proj.weight",
243
+ "shape": [
244
+ 3072,
245
+ 8192
246
+ ],
247
+ "dtype": "bfloat16",
248
+ "format": "raw",
249
+ "nbytes": 50331648,
250
+ "byteOffset": 0
251
+ }
252
+ ],
253
+ "md5sum": "8e2ea40c1f92cbb61b9d658895140b07"
254
+ },
255
+ {
256
+ "dataPath": "params_shard_13.bin",
257
+ "format": "raw-shard",
258
+ "nbytes": 100663296,
259
+ "records": [
260
+ {
261
+ "name": "model.layers.11.mlp.gate_up_proj.weight",
262
+ "shape": [
263
+ 16384,
264
+ 3072
265
+ ],
266
+ "dtype": "bfloat16",
267
+ "format": "raw",
268
+ "nbytes": 100663296,
269
+ "byteOffset": 0
270
+ }
271
+ ],
272
+ "md5sum": "083ec2e27ba602a63ebb32ce85aff2db"
273
+ },
274
+ {
275
+ "dataPath": "params_shard_14.bin",
276
+ "format": "raw-shard",
277
+ "nbytes": 31457280,
278
+ "records": [
279
+ {
280
+ "name": "model.layers.11.self_attn.qkv_proj.weight",
281
+ "shape": [
282
+ 5120,
283
+ 3072
284
+ ],
285
+ "dtype": "bfloat16",
286
+ "format": "raw",
287
+ "nbytes": 31457280,
288
+ "byteOffset": 0
289
+ }
290
+ ],
291
+ "md5sum": "94dd82b511ec123317e8ece8285765af"
292
+ },
293
+ {
294
+ "dataPath": "params_shard_15.bin",
295
+ "format": "raw-shard",
296
+ "nbytes": 18874368,
297
+ "records": [
298
+ {
299
+ "name": "model.layers.11.self_attn.o_proj.weight",
300
+ "shape": [
301
+ 3072,
302
+ 3072
303
+ ],
304
+ "dtype": "bfloat16",
305
+ "format": "raw",
306
+ "nbytes": 18874368,
307
+ "byteOffset": 0
308
+ }
309
+ ],
310
+ "md5sum": "75fc3f6b90b60616d2696b12412fc23b"
311
+ },
312
+ {
313
+ "dataPath": "params_shard_16.bin",
314
+ "format": "raw-shard",
315
+ "nbytes": 50331648,
316
+ "records": [
317
+ {
318
+ "name": "model.layers.12.mlp.down_proj.weight",
319
+ "shape": [
320
+ 3072,
321
+ 8192
322
+ ],
323
+ "dtype": "bfloat16",
324
+ "format": "raw",
325
+ "nbytes": 50331648,
326
+ "byteOffset": 0
327
+ }
328
+ ],
329
+ "md5sum": "1f5e4e0701c74744048f742778298f61"
330
+ },
331
+ {
332
+ "dataPath": "params_shard_17.bin",
333
+ "format": "raw-shard",
334
+ "nbytes": 100663296,
335
+ "records": [
336
+ {
337
+ "name": "model.layers.12.mlp.gate_up_proj.weight",
338
+ "shape": [
339
+ 16384,
340
+ 3072
341
+ ],
342
+ "dtype": "bfloat16",
343
+ "format": "raw",
344
+ "nbytes": 100663296,
345
+ "byteOffset": 0
346
+ }
347
+ ],
348
+ "md5sum": "91108c2df99c4b9b87f27edc7c75d600"
349
+ },
350
+ {
351
+ "dataPath": "params_shard_18.bin",
352
+ "format": "raw-shard",
353
+ "nbytes": 31457280,
354
+ "records": [
355
+ {
356
+ "name": "model.layers.12.self_attn.qkv_proj.weight",
357
+ "shape": [
358
+ 5120,
359
+ 3072
360
+ ],
361
+ "dtype": "bfloat16",
362
+ "format": "raw",
363
+ "nbytes": 31457280,
364
+ "byteOffset": 0
365
+ }
366
+ ],
367
+ "md5sum": "a8abdbceeb4dd058614f3f144a77ece2"
368
+ },
369
+ {
370
+ "dataPath": "params_shard_19.bin",
371
+ "format": "raw-shard",
372
+ "nbytes": 18874368,
373
+ "records": [
374
+ {
375
+ "name": "model.layers.12.self_attn.o_proj.weight",
376
+ "shape": [
377
+ 3072,
378
+ 3072
379
+ ],
380
+ "dtype": "bfloat16",
381
+ "format": "raw",
382
+ "nbytes": 18874368,
383
+ "byteOffset": 0
384
+ }
385
+ ],
386
+ "md5sum": "c75eab31f6da7c631a17e2ce852cf003"
387
+ },
388
+ {
389
+ "dataPath": "params_shard_20.bin",
390
+ "format": "raw-shard",
391
+ "nbytes": 50331648,
392
+ "records": [
393
+ {
394
+ "name": "model.layers.13.mlp.down_proj.weight",
395
+ "shape": [
396
+ 3072,
397
+ 8192
398
+ ],
399
+ "dtype": "bfloat16",
400
+ "format": "raw",
401
+ "nbytes": 50331648,
402
+ "byteOffset": 0
403
+ }
404
+ ],
405
+ "md5sum": "c6dbd24e7eb2d5485a7fa009326ca523"
406
+ },
407
+ {
408
+ "dataPath": "params_shard_21.bin",
409
+ "format": "raw-shard",
410
+ "nbytes": 100663296,
411
+ "records": [
412
+ {
413
+ "name": "model.layers.13.mlp.gate_up_proj.weight",
414
+ "shape": [
415
+ 16384,
416
+ 3072
417
+ ],
418
+ "dtype": "bfloat16",
419
+ "format": "raw",
420
+ "nbytes": 100663296,
421
+ "byteOffset": 0
422
+ }
423
+ ],
424
+ "md5sum": "ce2e8339c0cb6d4457a5ede2b3bee0b5"
425
+ },
426
+ {
427
+ "dataPath": "params_shard_22.bin",
428
+ "format": "raw-shard",
429
+ "nbytes": 31457280,
430
+ "records": [
431
+ {
432
+ "name": "model.layers.13.self_attn.qkv_proj.weight",
433
+ "shape": [
434
+ 5120,
435
+ 3072
436
+ ],
437
+ "dtype": "bfloat16",
438
+ "format": "raw",
439
+ "nbytes": 31457280,
440
+ "byteOffset": 0
441
+ }
442
+ ],
443
+ "md5sum": "611b51469fd096b07a9463d8f2a664eb"
444
+ },
445
+ {
446
+ "dataPath": "params_shard_23.bin",
447
+ "format": "raw-shard",
448
+ "nbytes": 18874368,
449
+ "records": [
450
+ {
451
+ "name": "model.layers.13.self_attn.o_proj.weight",
452
+ "shape": [
453
+ 3072,
454
+ 3072
455
+ ],
456
+ "dtype": "bfloat16",
457
+ "format": "raw",
458
+ "nbytes": 18874368,
459
+ "byteOffset": 0
460
+ }
461
+ ],
462
+ "md5sum": "50dc7e6b1fbee76eaa87ce0f5ec3fcc0"
463
+ },
464
+ {
465
+ "dataPath": "params_shard_24.bin",
466
+ "format": "raw-shard",
467
+ "nbytes": 50331648,
468
+ "records": [
469
+ {
470
+ "name": "model.layers.14.mlp.down_proj.weight",
471
+ "shape": [
472
+ 3072,
473
+ 8192
474
+ ],
475
+ "dtype": "bfloat16",
476
+ "format": "raw",
477
+ "nbytes": 50331648,
478
+ "byteOffset": 0
479
+ }
480
+ ],
481
+ "md5sum": "c6f7f3faeaef9e79991b7d94b25ee011"
482
+ },
483
+ {
484
+ "dataPath": "params_shard_25.bin",
485
+ "format": "raw-shard",
486
+ "nbytes": 100663296,
487
+ "records": [
488
+ {
489
+ "name": "model.layers.14.mlp.gate_up_proj.weight",
490
+ "shape": [
491
+ 16384,
492
+ 3072
493
+ ],
494
+ "dtype": "bfloat16",
495
+ "format": "raw",
496
+ "nbytes": 100663296,
497
+ "byteOffset": 0
498
+ }
499
+ ],
500
+ "md5sum": "815b14c6b06f48a87b5adc0cca956cfd"
501
+ },
502
+ {
503
+ "dataPath": "params_shard_26.bin",
504
+ "format": "raw-shard",
505
+ "nbytes": 31457280,
506
+ "records": [
507
+ {
508
+ "name": "model.layers.14.self_attn.qkv_proj.weight",
509
+ "shape": [
510
+ 5120,
511
+ 3072
512
+ ],
513
+ "dtype": "bfloat16",
514
+ "format": "raw",
515
+ "nbytes": 31457280,
516
+ "byteOffset": 0
517
+ }
518
+ ],
519
+ "md5sum": "0916f5bed827fa16ce1ebe9c089097d4"
520
+ },
521
+ {
522
+ "dataPath": "params_shard_27.bin",
523
+ "format": "raw-shard",
524
+ "nbytes": 18874368,
525
+ "records": [
526
+ {
527
+ "name": "model.layers.14.self_attn.o_proj.weight",
528
+ "shape": [
529
+ 3072,
530
+ 3072
531
+ ],
532
+ "dtype": "bfloat16",
533
+ "format": "raw",
534
+ "nbytes": 18874368,
535
+ "byteOffset": 0
536
+ }
537
+ ],
538
+ "md5sum": "9ddd05a218bf06c0868f80bd0b270eda"
539
+ },
540
+ {
541
+ "dataPath": "params_shard_28.bin",
542
+ "format": "raw-shard",
543
+ "nbytes": 50331648,
544
+ "records": [
545
+ {
546
+ "name": "model.layers.15.mlp.down_proj.weight",
547
+ "shape": [
548
+ 3072,
549
+ 8192
550
+ ],
551
+ "dtype": "bfloat16",
552
+ "format": "raw",
553
+ "nbytes": 50331648,
554
+ "byteOffset": 0
555
+ }
556
+ ],
557
+ "md5sum": "9e02113405c9b93e9eecb883f1db14d5"
558
+ },
559
+ {
560
+ "dataPath": "params_shard_29.bin",
561
+ "format": "raw-shard",
562
+ "nbytes": 100663296,
563
+ "records": [
564
+ {
565
+ "name": "model.layers.15.mlp.gate_up_proj.weight",
566
+ "shape": [
567
+ 16384,
568
+ 3072
569
+ ],
570
+ "dtype": "bfloat16",
571
+ "format": "raw",
572
+ "nbytes": 100663296,
573
+ "byteOffset": 0
574
+ }
575
+ ],
576
+ "md5sum": "303e6706d5782c66c3a11756ee513dbb"
577
+ },
578
+ {
579
+ "dataPath": "params_shard_30.bin",
580
+ "format": "raw-shard",
581
+ "nbytes": 31457280,
582
+ "records": [
583
+ {
584
+ "name": "model.layers.15.self_attn.qkv_proj.weight",
585
+ "shape": [
586
+ 5120,
587
+ 3072
588
+ ],
589
+ "dtype": "bfloat16",
590
+ "format": "raw",
591
+ "nbytes": 31457280,
592
+ "byteOffset": 0
593
+ }
594
+ ],
595
+ "md5sum": "59c43454b5e413cbaab867a90ef827f3"
596
+ },
597
+ {
598
+ "dataPath": "params_shard_31.bin",
599
+ "format": "raw-shard",
600
+ "nbytes": 18874368,
601
+ "records": [
602
+ {
603
+ "name": "model.layers.15.self_attn.o_proj.weight",
604
+ "shape": [
605
+ 3072,
606
+ 3072
607
+ ],
608
+ "dtype": "bfloat16",
609
+ "format": "raw",
610
+ "nbytes": 18874368,
611
+ "byteOffset": 0
612
+ }
613
+ ],
614
+ "md5sum": "f148aa81afde44c61e15ad4ddf01c8a9"
615
+ },
616
+ {
617
+ "dataPath": "params_shard_32.bin",
618
+ "format": "raw-shard",
619
+ "nbytes": 50331648,
620
+ "records": [
621
+ {
622
+ "name": "model.layers.16.mlp.down_proj.weight",
623
+ "shape": [
624
+ 3072,
625
+ 8192
626
+ ],
627
+ "dtype": "bfloat16",
628
+ "format": "raw",
629
+ "nbytes": 50331648,
630
+ "byteOffset": 0
631
+ }
632
+ ],
633
+ "md5sum": "c1bc09fcd5aedc98f925795d44bd597a"
634
+ },
635
+ {
636
+ "dataPath": "params_shard_33.bin",
637
+ "format": "raw-shard",
638
+ "nbytes": 100663296,
639
+ "records": [
640
+ {
641
+ "name": "model.layers.16.mlp.gate_up_proj.weight",
642
+ "shape": [
643
+ 16384,
644
+ 3072
645
+ ],
646
+ "dtype": "bfloat16",
647
+ "format": "raw",
648
+ "nbytes": 100663296,
649
+ "byteOffset": 0
650
+ }
651
+ ],
652
+ "md5sum": "d75aa45200b65bc44500abb95e9efdcc"
653
+ },
654
+ {
655
+ "dataPath": "params_shard_34.bin",
656
+ "format": "raw-shard",
657
+ "nbytes": 31457280,
658
+ "records": [
659
+ {
660
+ "name": "model.layers.16.self_attn.qkv_proj.weight",
661
+ "shape": [
662
+ 5120,
663
+ 3072
664
+ ],
665
+ "dtype": "bfloat16",
666
+ "format": "raw",
667
+ "nbytes": 31457280,
668
+ "byteOffset": 0
669
+ }
670
+ ],
671
+ "md5sum": "d2607a22d2aaa52916e1f99977674d3c"
672
+ },
673
+ {
674
+ "dataPath": "params_shard_35.bin",
675
+ "format": "raw-shard",
676
+ "nbytes": 18874368,
677
+ "records": [
678
+ {
679
+ "name": "model.layers.16.self_attn.o_proj.weight",
680
+ "shape": [
681
+ 3072,
682
+ 3072
683
+ ],
684
+ "dtype": "bfloat16",
685
+ "format": "raw",
686
+ "nbytes": 18874368,
687
+ "byteOffset": 0
688
+ }
689
+ ],
690
+ "md5sum": "6b0bc187e174276f9068c71f001861af"
691
+ },
692
+ {
693
+ "dataPath": "params_shard_36.bin",
694
+ "format": "raw-shard",
695
+ "nbytes": 50331648,
696
+ "records": [
697
+ {
698
+ "name": "model.layers.17.mlp.down_proj.weight",
699
+ "shape": [
700
+ 3072,
701
+ 8192
702
+ ],
703
+ "dtype": "bfloat16",
704
+ "format": "raw",
705
+ "nbytes": 50331648,
706
+ "byteOffset": 0
707
+ }
708
+ ],
709
+ "md5sum": "f29349e71e3c7cc0c5172fb8de9e637f"
710
+ },
711
+ {
712
+ "dataPath": "params_shard_37.bin",
713
+ "format": "raw-shard",
714
+ "nbytes": 100663296,
715
+ "records": [
716
+ {
717
+ "name": "model.layers.17.mlp.gate_up_proj.weight",
718
+ "shape": [
719
+ 16384,
720
+ 3072
721
+ ],
722
+ "dtype": "bfloat16",
723
+ "format": "raw",
724
+ "nbytes": 100663296,
725
+ "byteOffset": 0
726
+ }
727
+ ],
728
+ "md5sum": "b98a90817c9fdb702470209e341ddd5c"
729
+ },
730
+ {
731
+ "dataPath": "params_shard_38.bin",
732
+ "format": "raw-shard",
733
+ "nbytes": 31457280,
734
+ "records": [
735
+ {
736
+ "name": "model.layers.17.self_attn.qkv_proj.weight",
737
+ "shape": [
738
+ 5120,
739
+ 3072
740
+ ],
741
+ "dtype": "bfloat16",
742
+ "format": "raw",
743
+ "nbytes": 31457280,
744
+ "byteOffset": 0
745
+ }
746
+ ],
747
+ "md5sum": "56b6d3ede5ea23c9994fb02c48a82cd7"
748
+ },
749
+ {
750
+ "dataPath": "params_shard_39.bin",
751
+ "format": "raw-shard",
752
+ "nbytes": 18874368,
753
+ "records": [
754
+ {
755
+ "name": "model.layers.17.self_attn.o_proj.weight",
756
+ "shape": [
757
+ 3072,
758
+ 3072
759
+ ],
760
+ "dtype": "bfloat16",
761
+ "format": "raw",
762
+ "nbytes": 18874368,
763
+ "byteOffset": 0
764
+ }
765
+ ],
766
+ "md5sum": "f6c60380deb7a8afbed11b307b145e18"
767
+ },
768
+ {
769
+ "dataPath": "params_shard_40.bin",
770
+ "format": "raw-shard",
771
+ "nbytes": 50331648,
772
+ "records": [
773
+ {
774
+ "name": "model.layers.18.mlp.down_proj.weight",
775
+ "shape": [
776
+ 3072,
777
+ 8192
778
+ ],
779
+ "dtype": "bfloat16",
780
+ "format": "raw",
781
+ "nbytes": 50331648,
782
+ "byteOffset": 0
783
+ }
784
+ ],
785
+ "md5sum": "088c2b1caad88b346fbb5a3cb0d45110"
786
+ },
787
+ {
788
+ "dataPath": "params_shard_41.bin",
789
+ "format": "raw-shard",
790
+ "nbytes": 100663296,
791
+ "records": [
792
+ {
793
+ "name": "model.layers.18.mlp.gate_up_proj.weight",
794
+ "shape": [
795
+ 16384,
796
+ 3072
797
+ ],
798
+ "dtype": "bfloat16",
799
+ "format": "raw",
800
+ "nbytes": 100663296,
801
+ "byteOffset": 0
802
+ }
803
+ ],
804
+ "md5sum": "44d02bd8483c1a0941af26e32438bb55"
805
+ },
806
+ {
807
+ "dataPath": "params_shard_42.bin",
808
+ "format": "raw-shard",
809
+ "nbytes": 31457280,
810
+ "records": [
811
+ {
812
+ "name": "model.layers.18.self_attn.qkv_proj.weight",
813
+ "shape": [
814
+ 5120,
815
+ 3072
816
+ ],
817
+ "dtype": "bfloat16",
818
+ "format": "raw",
819
+ "nbytes": 31457280,
820
+ "byteOffset": 0
821
+ }
822
+ ],
823
+ "md5sum": "824146de15eb8206b9e7531a3a068b6f"
824
+ },
825
+ {
826
+ "dataPath": "params_shard_43.bin",
827
+ "format": "raw-shard",
828
+ "nbytes": 18874368,
829
+ "records": [
830
+ {
831
+ "name": "model.layers.18.self_attn.o_proj.weight",
832
+ "shape": [
833
+ 3072,
834
+ 3072
835
+ ],
836
+ "dtype": "bfloat16",
837
+ "format": "raw",
838
+ "nbytes": 18874368,
839
+ "byteOffset": 0
840
+ }
841
+ ],
842
+ "md5sum": "ad1677d7e01e4ef22f342607f16f02df"
843
+ },
844
+ {
845
+ "dataPath": "params_shard_44.bin",
846
+ "format": "raw-shard",
847
+ "nbytes": 50331648,
848
+ "records": [
849
+ {
850
+ "name": "model.layers.19.mlp.down_proj.weight",
851
+ "shape": [
852
+ 3072,
853
+ 8192
854
+ ],
855
+ "dtype": "bfloat16",
856
+ "format": "raw",
857
+ "nbytes": 50331648,
858
+ "byteOffset": 0
859
+ }
860
+ ],
861
+ "md5sum": "3143003000df5244f9fcfda00b62e2a0"
862
+ },
863
+ {
864
+ "dataPath": "params_shard_45.bin",
865
+ "format": "raw-shard",
866
+ "nbytes": 100663296,
867
+ "records": [
868
+ {
869
+ "name": "model.layers.19.mlp.gate_up_proj.weight",
870
+ "shape": [
871
+ 16384,
872
+ 3072
873
+ ],
874
+ "dtype": "bfloat16",
875
+ "format": "raw",
876
+ "nbytes": 100663296,
877
+ "byteOffset": 0
878
+ }
879
+ ],
880
+ "md5sum": "96d5a23e28e0cd687ee129bf25bdd674"
881
+ },
882
+ {
883
+ "dataPath": "params_shard_46.bin",
884
+ "format": "raw-shard",
885
+ "nbytes": 31457280,
886
+ "records": [
887
+ {
888
+ "name": "model.layers.19.self_attn.qkv_proj.weight",
889
+ "shape": [
890
+ 5120,
891
+ 3072
892
+ ],
893
+ "dtype": "bfloat16",
894
+ "format": "raw",
895
+ "nbytes": 31457280,
896
+ "byteOffset": 0
897
+ }
898
+ ],
899
+ "md5sum": "e7445744de4fce67eecd628cb5955f42"
900
+ },
901
+ {
902
+ "dataPath": "params_shard_47.bin",
903
+ "format": "raw-shard",
904
+ "nbytes": 18874368,
905
+ "records": [
906
+ {
907
+ "name": "model.layers.19.self_attn.o_proj.weight",
908
+ "shape": [
909
+ 3072,
910
+ 3072
911
+ ],
912
+ "dtype": "bfloat16",
913
+ "format": "raw",
914
+ "nbytes": 18874368,
915
+ "byteOffset": 0
916
+ }
917
+ ],
918
+ "md5sum": "4e8a3a848bc7c572129bde24545e91ab"
919
+ },
920
+ {
921
+ "dataPath": "params_shard_48.bin",
922
+ "format": "raw-shard",
923
+ "nbytes": 50331648,
924
+ "records": [
925
+ {
926
+ "name": "model.layers.2.mlp.down_proj.weight",
927
+ "shape": [
928
+ 3072,
929
+ 8192
930
+ ],
931
+ "dtype": "bfloat16",
932
+ "format": "raw",
933
+ "nbytes": 50331648,
934
+ "byteOffset": 0
935
+ }
936
+ ],
937
+ "md5sum": "c0d47846f1f268c6b9d034420139e267"
938
+ },
939
+ {
940
+ "dataPath": "params_shard_49.bin",
941
+ "format": "raw-shard",
942
+ "nbytes": 100663296,
943
+ "records": [
944
+ {
945
+ "name": "model.layers.2.mlp.gate_up_proj.weight",
946
+ "shape": [
947
+ 16384,
948
+ 3072
949
+ ],
950
+ "dtype": "bfloat16",
951
+ "format": "raw",
952
+ "nbytes": 100663296,
953
+ "byteOffset": 0
954
+ }
955
+ ],
956
+ "md5sum": "ff5c419b4b570dfa26fe530cd0602a75"
957
+ },
958
+ {
959
+ "dataPath": "params_shard_50.bin",
960
+ "format": "raw-shard",
961
+ "nbytes": 31457280,
962
+ "records": [
963
+ {
964
+ "name": "model.layers.2.self_attn.qkv_proj.weight",
965
+ "shape": [
966
+ 5120,
967
+ 3072
968
+ ],
969
+ "dtype": "bfloat16",
970
+ "format": "raw",
971
+ "nbytes": 31457280,
972
+ "byteOffset": 0
973
+ }
974
+ ],
975
+ "md5sum": "c11179f8247d6d4c364971a6450740a4"
976
+ },
977
+ {
978
+ "dataPath": "params_shard_51.bin",
979
+ "format": "raw-shard",
980
+ "nbytes": 18874368,
981
+ "records": [
982
+ {
983
+ "name": "model.layers.2.self_attn.o_proj.weight",
984
+ "shape": [
985
+ 3072,
986
+ 3072
987
+ ],
988
+ "dtype": "bfloat16",
989
+ "format": "raw",
990
+ "nbytes": 18874368,
991
+ "byteOffset": 0
992
+ }
993
+ ],
994
+ "md5sum": "7588e5e7fa4304beb826fcaa6f3b3e65"
995
+ },
996
+ {
997
+ "dataPath": "params_shard_52.bin",
998
+ "format": "raw-shard",
999
+ "nbytes": 100663296,
1000
+ "records": [
1001
+ {
1002
+ "name": "model.layers.20.mlp.gate_up_proj.weight",
1003
+ "shape": [
1004
+ 16384,
1005
+ 3072
1006
+ ],
1007
+ "dtype": "bfloat16",
1008
+ "format": "raw",
1009
+ "nbytes": 100663296,
1010
+ "byteOffset": 0
1011
+ }
1012
+ ],
1013
+ "md5sum": "49d7128811effec28ba14cd181946cb9"
1014
+ },
1015
+ {
1016
+ "dataPath": "params_shard_53.bin",
1017
+ "format": "raw-shard",
1018
+ "nbytes": 31457280,
1019
+ "records": [
1020
+ {
1021
+ "name": "model.layers.20.self_attn.qkv_proj.weight",
1022
+ "shape": [
1023
+ 5120,
1024
+ 3072
1025
+ ],
1026
+ "dtype": "bfloat16",
1027
+ "format": "raw",
1028
+ "nbytes": 31457280,
1029
+ "byteOffset": 0
1030
+ }
1031
+ ],
1032
+ "md5sum": "b2559fc7223d8081ea4ff1f38efe170d"
1033
+ },
1034
+ {
1035
+ "dataPath": "params_shard_54.bin",
1036
+ "format": "raw-shard",
1037
+ "nbytes": 18874368,
1038
+ "records": [
1039
+ {
1040
+ "name": "model.layers.20.self_attn.o_proj.weight",
1041
+ "shape": [
1042
+ 3072,
1043
+ 3072
1044
+ ],
1045
+ "dtype": "bfloat16",
1046
+ "format": "raw",
1047
+ "nbytes": 18874368,
1048
+ "byteOffset": 0
1049
+ }
1050
+ ],
1051
+ "md5sum": "427dbed3aa91d806dfcb213501ac507e"
1052
+ },
1053
+ {
1054
+ "dataPath": "params_shard_55.bin",
1055
+ "format": "raw-shard",
1056
+ "nbytes": 50331648,
1057
+ "records": [
1058
+ {
1059
+ "name": "model.layers.3.mlp.down_proj.weight",
1060
+ "shape": [
1061
+ 3072,
1062
+ 8192
1063
+ ],
1064
+ "dtype": "bfloat16",
1065
+ "format": "raw",
1066
+ "nbytes": 50331648,
1067
+ "byteOffset": 0
1068
+ }
1069
+ ],
1070
+ "md5sum": "c86c1149c905eeca37abb6ace6b65081"
1071
+ },
1072
+ {
1073
+ "dataPath": "params_shard_56.bin",
1074
+ "format": "raw-shard",
1075
+ "nbytes": 100663296,
1076
+ "records": [
1077
+ {
1078
+ "name": "model.layers.3.mlp.gate_up_proj.weight",
1079
+ "shape": [
1080
+ 16384,
1081
+ 3072
1082
+ ],
1083
+ "dtype": "bfloat16",
1084
+ "format": "raw",
1085
+ "nbytes": 100663296,
1086
+ "byteOffset": 0
1087
+ }
1088
+ ],
1089
+ "md5sum": "d0270175e869b30212326d8898f6bc6e"
1090
+ },
1091
+ {
1092
+ "dataPath": "params_shard_57.bin",
1093
+ "format": "raw-shard",
1094
+ "nbytes": 31457280,
1095
+ "records": [
1096
+ {
1097
+ "name": "model.layers.3.self_attn.qkv_proj.weight",
1098
+ "shape": [
1099
+ 5120,
1100
+ 3072
1101
+ ],
1102
+ "dtype": "bfloat16",
1103
+ "format": "raw",
1104
+ "nbytes": 31457280,
1105
+ "byteOffset": 0
1106
+ }
1107
+ ],
1108
+ "md5sum": "97edc380dd2ad36b07b1dc01832a4407"
1109
+ },
1110
+ {
1111
+ "dataPath": "params_shard_58.bin",
1112
+ "format": "raw-shard",
1113
+ "nbytes": 18874368,
1114
+ "records": [
1115
+ {
1116
+ "name": "model.layers.3.self_attn.o_proj.weight",
1117
+ "shape": [
1118
+ 3072,
1119
+ 3072
1120
+ ],
1121
+ "dtype": "bfloat16",
1122
+ "format": "raw",
1123
+ "nbytes": 18874368,
1124
+ "byteOffset": 0
1125
+ }
1126
+ ],
1127
+ "md5sum": "517ca02bf6261842f24247640bc0903f"
1128
+ },
1129
+ {
1130
+ "dataPath": "params_shard_59.bin",
1131
+ "format": "raw-shard",
1132
+ "nbytes": 50331648,
1133
+ "records": [
1134
+ {
1135
+ "name": "model.layers.4.mlp.down_proj.weight",
1136
+ "shape": [
1137
+ 3072,
1138
+ 8192
1139
+ ],
1140
+ "dtype": "bfloat16",
1141
+ "format": "raw",
1142
+ "nbytes": 50331648,
1143
+ "byteOffset": 0
1144
+ }
1145
+ ],
1146
+ "md5sum": "7ae5988d6a252fc160f8afec618d78f8"
1147
+ },
1148
+ {
1149
+ "dataPath": "params_shard_60.bin",
1150
+ "format": "raw-shard",
1151
+ "nbytes": 100663296,
1152
+ "records": [
1153
+ {
1154
+ "name": "model.layers.4.mlp.gate_up_proj.weight",
1155
+ "shape": [
1156
+ 16384,
1157
+ 3072
1158
+ ],
1159
+ "dtype": "bfloat16",
1160
+ "format": "raw",
1161
+ "nbytes": 100663296,
1162
+ "byteOffset": 0
1163
+ }
1164
+ ],
1165
+ "md5sum": "31d045e4bc48359a0a9355a6b1227faa"
1166
+ },
1167
+ {
1168
+ "dataPath": "params_shard_61.bin",
1169
+ "format": "raw-shard",
1170
+ "nbytes": 31457280,
1171
+ "records": [
1172
+ {
1173
+ "name": "model.layers.4.self_attn.qkv_proj.weight",
1174
+ "shape": [
1175
+ 5120,
1176
+ 3072
1177
+ ],
1178
+ "dtype": "bfloat16",
1179
+ "format": "raw",
1180
+ "nbytes": 31457280,
1181
+ "byteOffset": 0
1182
+ }
1183
+ ],
1184
+ "md5sum": "2499d76e5b76de9a57ee472ec90d7385"
1185
+ },
1186
+ {
1187
+ "dataPath": "params_shard_62.bin",
1188
+ "format": "raw-shard",
1189
+ "nbytes": 18874368,
1190
+ "records": [
1191
+ {
1192
+ "name": "model.layers.4.self_attn.o_proj.weight",
1193
+ "shape": [
1194
+ 3072,
1195
+ 3072
1196
+ ],
1197
+ "dtype": "bfloat16",
1198
+ "format": "raw",
1199
+ "nbytes": 18874368,
1200
+ "byteOffset": 0
1201
+ }
1202
+ ],
1203
+ "md5sum": "75c244d8197b1919723016fa125500b5"
1204
+ },
1205
+ {
1206
+ "dataPath": "params_shard_63.bin",
1207
+ "format": "raw-shard",
1208
+ "nbytes": 50331648,
1209
+ "records": [
1210
+ {
1211
+ "name": "model.layers.5.mlp.down_proj.weight",
1212
+ "shape": [
1213
+ 3072,
1214
+ 8192
1215
+ ],
1216
+ "dtype": "bfloat16",
1217
+ "format": "raw",
1218
+ "nbytes": 50331648,
1219
+ "byteOffset": 0
1220
+ }
1221
+ ],
1222
+ "md5sum": "33c9ea14ac77d2ece67498ce22af6419"
1223
+ },
1224
+ {
1225
+ "dataPath": "params_shard_64.bin",
1226
+ "format": "raw-shard",
1227
+ "nbytes": 100663296,
1228
+ "records": [
1229
+ {
1230
+ "name": "model.layers.5.mlp.gate_up_proj.weight",
1231
+ "shape": [
1232
+ 16384,
1233
+ 3072
1234
+ ],
1235
+ "dtype": "bfloat16",
1236
+ "format": "raw",
1237
+ "nbytes": 100663296,
1238
+ "byteOffset": 0
1239
+ }
1240
+ ],
1241
+ "md5sum": "5f901b9695815d535efed5b0a4a1e931"
1242
+ },
1243
+ {
1244
+ "dataPath": "params_shard_65.bin",
1245
+ "format": "raw-shard",
1246
+ "nbytes": 31457280,
1247
+ "records": [
1248
+ {
1249
+ "name": "model.layers.5.self_attn.qkv_proj.weight",
1250
+ "shape": [
1251
+ 5120,
1252
+ 3072
1253
+ ],
1254
+ "dtype": "bfloat16",
1255
+ "format": "raw",
1256
+ "nbytes": 31457280,
1257
+ "byteOffset": 0
1258
+ }
1259
+ ],
1260
+ "md5sum": "d7806491318f8baaf9769a0bded6dc0b"
1261
+ },
1262
+ {
1263
+ "dataPath": "params_shard_66.bin",
1264
+ "format": "raw-shard",
1265
+ "nbytes": 18874368,
1266
+ "records": [
1267
+ {
1268
+ "name": "model.layers.5.self_attn.o_proj.weight",
1269
+ "shape": [
1270
+ 3072,
1271
+ 3072
1272
+ ],
1273
+ "dtype": "bfloat16",
1274
+ "format": "raw",
1275
+ "nbytes": 18874368,
1276
+ "byteOffset": 0
1277
+ }
1278
+ ],
1279
+ "md5sum": "7d60368414ca9cab75b2ae03b15fb208"
1280
+ },
1281
+ {
1282
+ "dataPath": "params_shard_67.bin",
1283
+ "format": "raw-shard",
1284
+ "nbytes": 50331648,
1285
+ "records": [
1286
+ {
1287
+ "name": "model.layers.6.mlp.down_proj.weight",
1288
+ "shape": [
1289
+ 3072,
1290
+ 8192
1291
+ ],
1292
+ "dtype": "bfloat16",
1293
+ "format": "raw",
1294
+ "nbytes": 50331648,
1295
+ "byteOffset": 0
1296
+ }
1297
+ ],
1298
+ "md5sum": "3052bcd3ac2b8d80ef17287f901bb3b9"
1299
+ },
1300
+ {
1301
+ "dataPath": "params_shard_68.bin",
1302
+ "format": "raw-shard",
1303
+ "nbytes": 100663296,
1304
+ "records": [
1305
+ {
1306
+ "name": "model.layers.6.mlp.gate_up_proj.weight",
1307
+ "shape": [
1308
+ 16384,
1309
+ 3072
1310
+ ],
1311
+ "dtype": "bfloat16",
1312
+ "format": "raw",
1313
+ "nbytes": 100663296,
1314
+ "byteOffset": 0
1315
+ }
1316
+ ],
1317
+ "md5sum": "c511a17ee7127ea3514af602b5b5f75e"
1318
+ },
1319
+ {
1320
+ "dataPath": "params_shard_69.bin",
1321
+ "format": "raw-shard",
1322
+ "nbytes": 31457280,
1323
+ "records": [
1324
+ {
1325
+ "name": "model.layers.6.self_attn.qkv_proj.weight",
1326
+ "shape": [
1327
+ 5120,
1328
+ 3072
1329
+ ],
1330
+ "dtype": "bfloat16",
1331
+ "format": "raw",
1332
+ "nbytes": 31457280,
1333
+ "byteOffset": 0
1334
+ }
1335
+ ],
1336
+ "md5sum": "e6e79670e5e719b3cec1e97f318218df"
1337
+ },
1338
+ {
1339
+ "dataPath": "params_shard_70.bin",
1340
+ "format": "raw-shard",
1341
+ "nbytes": 18874368,
1342
+ "records": [
1343
+ {
1344
+ "name": "model.layers.6.self_attn.o_proj.weight",
1345
+ "shape": [
1346
+ 3072,
1347
+ 3072
1348
+ ],
1349
+ "dtype": "bfloat16",
1350
+ "format": "raw",
1351
+ "nbytes": 18874368,
1352
+ "byteOffset": 0
1353
+ }
1354
+ ],
1355
+ "md5sum": "90bb59b9defd080833ae995f838e39fc"
1356
+ },
1357
+ {
1358
+ "dataPath": "params_shard_71.bin",
1359
+ "format": "raw-shard",
1360
+ "nbytes": 50331648,
1361
+ "records": [
1362
+ {
1363
+ "name": "model.layers.7.mlp.down_proj.weight",
1364
+ "shape": [
1365
+ 3072,
1366
+ 8192
1367
+ ],
1368
+ "dtype": "bfloat16",
1369
+ "format": "raw",
1370
+ "nbytes": 50331648,
1371
+ "byteOffset": 0
1372
+ }
1373
+ ],
1374
+ "md5sum": "6f7406fcbbaf37055796ee5282acd5e8"
1375
+ },
1376
+ {
1377
+ "dataPath": "params_shard_72.bin",
1378
+ "format": "raw-shard",
1379
+ "nbytes": 100663296,
1380
+ "records": [
1381
+ {
1382
+ "name": "model.layers.7.mlp.gate_up_proj.weight",
1383
+ "shape": [
1384
+ 16384,
1385
+ 3072
1386
+ ],
1387
+ "dtype": "bfloat16",
1388
+ "format": "raw",
1389
+ "nbytes": 100663296,
1390
+ "byteOffset": 0
1391
+ }
1392
+ ],
1393
+ "md5sum": "d976299fe65efbcda8040d0bb0cdcff4"
1394
+ },
1395
+ {
1396
+ "dataPath": "params_shard_73.bin",
1397
+ "format": "raw-shard",
1398
+ "nbytes": 31457280,
1399
+ "records": [
1400
+ {
1401
+ "name": "model.layers.7.self_attn.qkv_proj.weight",
1402
+ "shape": [
1403
+ 5120,
1404
+ 3072
1405
+ ],
1406
+ "dtype": "bfloat16",
1407
+ "format": "raw",
1408
+ "nbytes": 31457280,
1409
+ "byteOffset": 0
1410
+ }
1411
+ ],
1412
+ "md5sum": "1fcb92f405a175f578814f4f4c1ad88a"
1413
+ },
1414
+ {
1415
+ "dataPath": "params_shard_74.bin",
1416
+ "format": "raw-shard",
1417
+ "nbytes": 18874368,
1418
+ "records": [
1419
+ {
1420
+ "name": "model.layers.7.self_attn.o_proj.weight",
1421
+ "shape": [
1422
+ 3072,
1423
+ 3072
1424
+ ],
1425
+ "dtype": "bfloat16",
1426
+ "format": "raw",
1427
+ "nbytes": 18874368,
1428
+ "byteOffset": 0
1429
+ }
1430
+ ],
1431
+ "md5sum": "71663fa01f184a407c2a5e19ec85e9e7"
1432
+ },
1433
+ {
1434
+ "dataPath": "params_shard_75.bin",
1435
+ "format": "raw-shard",
1436
+ "nbytes": 50331648,
1437
+ "records": [
1438
+ {
1439
+ "name": "model.layers.8.mlp.down_proj.weight",
1440
+ "shape": [
1441
+ 3072,
1442
+ 8192
1443
+ ],
1444
+ "dtype": "bfloat16",
1445
+ "format": "raw",
1446
+ "nbytes": 50331648,
1447
+ "byteOffset": 0
1448
+ }
1449
+ ],
1450
+ "md5sum": "4e9ba681099e252c6cd123007b45e416"
1451
+ },
1452
+ {
1453
+ "dataPath": "params_shard_76.bin",
1454
+ "format": "raw-shard",
1455
+ "nbytes": 100663296,
1456
+ "records": [
1457
+ {
1458
+ "name": "model.layers.8.mlp.gate_up_proj.weight",
1459
+ "shape": [
1460
+ 16384,
1461
+ 3072
1462
+ ],
1463
+ "dtype": "bfloat16",
1464
+ "format": "raw",
1465
+ "nbytes": 100663296,
1466
+ "byteOffset": 0
1467
+ }
1468
+ ],
1469
+ "md5sum": "e58dad56d1aa5aa3306cb3190ba7ff1f"
1470
+ },
1471
+ {
1472
+ "dataPath": "params_shard_77.bin",
1473
+ "format": "raw-shard",
1474
+ "nbytes": 31457280,
1475
+ "records": [
1476
+ {
1477
+ "name": "model.layers.8.self_attn.qkv_proj.weight",
1478
+ "shape": [
1479
+ 5120,
1480
+ 3072
1481
+ ],
1482
+ "dtype": "bfloat16",
1483
+ "format": "raw",
1484
+ "nbytes": 31457280,
1485
+ "byteOffset": 0
1486
+ }
1487
+ ],
1488
+ "md5sum": "0001d2ba2d5eea1354aef2f5c727fe07"
1489
+ },
1490
+ {
1491
+ "dataPath": "params_shard_78.bin",
1492
+ "format": "raw-shard",
1493
+ "nbytes": 18874368,
1494
+ "records": [
1495
+ {
1496
+ "name": "model.layers.8.self_attn.o_proj.weight",
1497
+ "shape": [
1498
+ 3072,
1499
+ 3072
1500
+ ],
1501
+ "dtype": "bfloat16",
1502
+ "format": "raw",
1503
+ "nbytes": 18874368,
1504
+ "byteOffset": 0
1505
+ }
1506
+ ],
1507
+ "md5sum": "76f9235ba682cf5b0388c94cdef2ec15"
1508
+ },
1509
+ {
1510
+ "dataPath": "params_shard_79.bin",
1511
+ "format": "raw-shard",
1512
+ "nbytes": 50331648,
1513
+ "records": [
1514
+ {
1515
+ "name": "model.layers.9.mlp.down_proj.weight",
1516
+ "shape": [
1517
+ 3072,
1518
+ 8192
1519
+ ],
1520
+ "dtype": "bfloat16",
1521
+ "format": "raw",
1522
+ "nbytes": 50331648,
1523
+ "byteOffset": 0
1524
+ }
1525
+ ],
1526
+ "md5sum": "5b7ac8c47930806d061094840999d5d7"
1527
+ },
1528
+ {
1529
+ "dataPath": "params_shard_80.bin",
1530
+ "format": "raw-shard",
1531
+ "nbytes": 100663296,
1532
+ "records": [
1533
+ {
1534
+ "name": "model.layers.9.mlp.gate_up_proj.weight",
1535
+ "shape": [
1536
+ 16384,
1537
+ 3072
1538
+ ],
1539
+ "dtype": "bfloat16",
1540
+ "format": "raw",
1541
+ "nbytes": 100663296,
1542
+ "byteOffset": 0
1543
+ }
1544
+ ],
1545
+ "md5sum": "02e74665c93354e4829fb04653abc27b"
1546
+ },
1547
+ {
1548
+ "dataPath": "params_shard_81.bin",
1549
+ "format": "raw-shard",
1550
+ "nbytes": 31457280,
1551
+ "records": [
1552
+ {
1553
+ "name": "model.layers.9.self_attn.qkv_proj.weight",
1554
+ "shape": [
1555
+ 5120,
1556
+ 3072
1557
+ ],
1558
+ "dtype": "bfloat16",
1559
+ "format": "raw",
1560
+ "nbytes": 31457280,
1561
+ "byteOffset": 0
1562
+ }
1563
+ ],
1564
+ "md5sum": "ac1cff11d9b4cc142961891e4b0fb7de"
1565
+ },
1566
+ {
1567
+ "dataPath": "params_shard_82.bin",
1568
+ "format": "raw-shard",
1569
+ "nbytes": 18874368,
1570
+ "records": [
1571
+ {
1572
+ "name": "model.layers.9.self_attn.o_proj.weight",
1573
+ "shape": [
1574
+ 3072,
1575
+ 3072
1576
+ ],
1577
+ "dtype": "bfloat16",
1578
+ "format": "raw",
1579
+ "nbytes": 18874368,
1580
+ "byteOffset": 0
1581
+ }
1582
+ ],
1583
+ "md5sum": "67bbc168157c53d1a1a1ff8b6052eaa1"
1584
+ },
1585
+ {
1586
+ "dataPath": "params_shard_83.bin",
1587
+ "format": "raw-shard",
1588
+ "nbytes": 50331648,
1589
+ "records": [
1590
+ {
1591
+ "name": "model.layers.20.mlp.down_proj.weight",
1592
+ "shape": [
1593
+ 3072,
1594
+ 8192
1595
+ ],
1596
+ "dtype": "bfloat16",
1597
+ "format": "raw",
1598
+ "nbytes": 50331648,
1599
+ "byteOffset": 0
1600
+ }
1601
+ ],
1602
+ "md5sum": "cef4f79d43d9eace1ad5c2bc0bbab32c"
1603
+ },
1604
+ {
1605
+ "dataPath": "params_shard_84.bin",
1606
+ "format": "raw-shard",
1607
+ "nbytes": 50331648,
1608
+ "records": [
1609
+ {
1610
+ "name": "model.layers.21.mlp.down_proj.weight",
1611
+ "shape": [
1612
+ 3072,
1613
+ 8192
1614
+ ],
1615
+ "dtype": "bfloat16",
1616
+ "format": "raw",
1617
+ "nbytes": 50331648,
1618
+ "byteOffset": 0
1619
+ }
1620
+ ],
1621
+ "md5sum": "caa538f23db663ddb35a0f7613bced7a"
1622
+ },
1623
+ {
1624
+ "dataPath": "params_shard_85.bin",
1625
+ "format": "raw-shard",
1626
+ "nbytes": 100663296,
1627
+ "records": [
1628
+ {
1629
+ "name": "model.layers.21.mlp.gate_up_proj.weight",
1630
+ "shape": [
1631
+ 16384,
1632
+ 3072
1633
+ ],
1634
+ "dtype": "bfloat16",
1635
+ "format": "raw",
1636
+ "nbytes": 100663296,
1637
+ "byteOffset": 0
1638
+ }
1639
+ ],
1640
+ "md5sum": "ae2c3067f8024380c9729aeb23055920"
1641
+ },
1642
+ {
1643
+ "dataPath": "params_shard_86.bin",
1644
+ "format": "raw-shard",
1645
+ "nbytes": 31457280,
1646
+ "records": [
1647
+ {
1648
+ "name": "model.layers.21.self_attn.qkv_proj.weight",
1649
+ "shape": [
1650
+ 5120,
1651
+ 3072
1652
+ ],
1653
+ "dtype": "bfloat16",
1654
+ "format": "raw",
1655
+ "nbytes": 31457280,
1656
+ "byteOffset": 0
1657
+ }
1658
+ ],
1659
+ "md5sum": "80fa2ae93a50724674d3ccf3097bc110"
1660
+ },
1661
+ {
1662
+ "dataPath": "params_shard_87.bin",
1663
+ "format": "raw-shard",
1664
+ "nbytes": 18874368,
1665
+ "records": [
1666
+ {
1667
+ "name": "model.layers.21.self_attn.o_proj.weight",
1668
+ "shape": [
1669
+ 3072,
1670
+ 3072
1671
+ ],
1672
+ "dtype": "bfloat16",
1673
+ "format": "raw",
1674
+ "nbytes": 18874368,
1675
+ "byteOffset": 0
1676
+ }
1677
+ ],
1678
+ "md5sum": "8005bb547cfd8eed05260de00e01d305"
1679
+ },
1680
+ {
1681
+ "dataPath": "params_shard_88.bin",
1682
+ "format": "raw-shard",
1683
+ "nbytes": 50331648,
1684
+ "records": [
1685
+ {
1686
+ "name": "model.layers.22.mlp.down_proj.weight",
1687
+ "shape": [
1688
+ 3072,
1689
+ 8192
1690
+ ],
1691
+ "dtype": "bfloat16",
1692
+ "format": "raw",
1693
+ "nbytes": 50331648,
1694
+ "byteOffset": 0
1695
+ }
1696
+ ],
1697
+ "md5sum": "55d1db437255049a98db5ca1d677dd25"
1698
+ },
1699
+ {
1700
+ "dataPath": "params_shard_89.bin",
1701
+ "format": "raw-shard",
1702
+ "nbytes": 100663296,
1703
+ "records": [
1704
+ {
1705
+ "name": "model.layers.22.mlp.gate_up_proj.weight",
1706
+ "shape": [
1707
+ 16384,
1708
+ 3072
1709
+ ],
1710
+ "dtype": "bfloat16",
1711
+ "format": "raw",
1712
+ "nbytes": 100663296,
1713
+ "byteOffset": 0
1714
+ }
1715
+ ],
1716
+ "md5sum": "e7f1f30f38670d683112ba5706a2f092"
1717
+ },
1718
+ {
1719
+ "dataPath": "params_shard_90.bin",
1720
+ "format": "raw-shard",
1721
+ "nbytes": 31457280,
1722
+ "records": [
1723
+ {
1724
+ "name": "model.layers.22.self_attn.qkv_proj.weight",
1725
+ "shape": [
1726
+ 5120,
1727
+ 3072
1728
+ ],
1729
+ "dtype": "bfloat16",
1730
+ "format": "raw",
1731
+ "nbytes": 31457280,
1732
+ "byteOffset": 0
1733
+ }
1734
+ ],
1735
+ "md5sum": "04a985de9da66349a9e0401546d9a065"
1736
+ },
1737
+ {
1738
+ "dataPath": "params_shard_91.bin",
1739
+ "format": "raw-shard",
1740
+ "nbytes": 18874368,
1741
+ "records": [
1742
+ {
1743
+ "name": "model.layers.22.self_attn.o_proj.weight",
1744
+ "shape": [
1745
+ 3072,
1746
+ 3072
1747
+ ],
1748
+ "dtype": "bfloat16",
1749
+ "format": "raw",
1750
+ "nbytes": 18874368,
1751
+ "byteOffset": 0
1752
+ }
1753
+ ],
1754
+ "md5sum": "65ecf4f35052edf63f9f3d20c470c5dd"
1755
+ },
1756
+ {
1757
+ "dataPath": "params_shard_92.bin",
1758
+ "format": "raw-shard",
1759
+ "nbytes": 50331648,
1760
+ "records": [
1761
+ {
1762
+ "name": "model.layers.23.mlp.down_proj.weight",
1763
+ "shape": [
1764
+ 3072,
1765
+ 8192
1766
+ ],
1767
+ "dtype": "bfloat16",
1768
+ "format": "raw",
1769
+ "nbytes": 50331648,
1770
+ "byteOffset": 0
1771
+ }
1772
+ ],
1773
+ "md5sum": "f0d1c8ba2a582fdcc0e5270f7c061471"
1774
+ },
1775
+ {
1776
+ "dataPath": "params_shard_93.bin",
1777
+ "format": "raw-shard",
1778
+ "nbytes": 100663296,
1779
+ "records": [
1780
+ {
1781
+ "name": "model.layers.23.mlp.gate_up_proj.weight",
1782
+ "shape": [
1783
+ 16384,
1784
+ 3072
1785
+ ],
1786
+ "dtype": "bfloat16",
1787
+ "format": "raw",
1788
+ "nbytes": 100663296,
1789
+ "byteOffset": 0
1790
+ }
1791
+ ],
1792
+ "md5sum": "a0a933b90f40ea9b50fdcde46463cd55"
1793
+ },
1794
+ {
1795
+ "dataPath": "params_shard_94.bin",
1796
+ "format": "raw-shard",
1797
+ "nbytes": 31457280,
1798
+ "records": [
1799
+ {
1800
+ "name": "model.layers.23.self_attn.qkv_proj.weight",
1801
+ "shape": [
1802
+ 5120,
1803
+ 3072
1804
+ ],
1805
+ "dtype": "bfloat16",
1806
+ "format": "raw",
1807
+ "nbytes": 31457280,
1808
+ "byteOffset": 0
1809
+ }
1810
+ ],
1811
+ "md5sum": "da99b4fb689c104ee62ad2cad0cfa534"
1812
+ },
1813
+ {
1814
+ "dataPath": "params_shard_95.bin",
1815
+ "format": "raw-shard",
1816
+ "nbytes": 18874368,
1817
+ "records": [
1818
+ {
1819
+ "name": "model.layers.23.self_attn.o_proj.weight",
1820
+ "shape": [
1821
+ 3072,
1822
+ 3072
1823
+ ],
1824
+ "dtype": "bfloat16",
1825
+ "format": "raw",
1826
+ "nbytes": 18874368,
1827
+ "byteOffset": 0
1828
+ }
1829
+ ],
1830
+ "md5sum": "0208161744636162aaddd958458c42f4"
1831
+ },
1832
+ {
1833
+ "dataPath": "params_shard_96.bin",
1834
+ "format": "raw-shard",
1835
+ "nbytes": 50331648,
1836
+ "records": [
1837
+ {
1838
+ "name": "model.layers.24.mlp.down_proj.weight",
1839
+ "shape": [
1840
+ 3072,
1841
+ 8192
1842
+ ],
1843
+ "dtype": "bfloat16",
1844
+ "format": "raw",
1845
+ "nbytes": 50331648,
1846
+ "byteOffset": 0
1847
+ }
1848
+ ],
1849
+ "md5sum": "1c87befe152239ad5a565da807630695"
1850
+ },
1851
+ {
1852
+ "dataPath": "params_shard_97.bin",
1853
+ "format": "raw-shard",
1854
+ "nbytes": 100663296,
1855
+ "records": [
1856
+ {
1857
+ "name": "model.layers.24.mlp.gate_up_proj.weight",
1858
+ "shape": [
1859
+ 16384,
1860
+ 3072
1861
+ ],
1862
+ "dtype": "bfloat16",
1863
+ "format": "raw",
1864
+ "nbytes": 100663296,
1865
+ "byteOffset": 0
1866
+ }
1867
+ ],
1868
+ "md5sum": "f9a41904bae1070952eabe69083f82d1"
1869
+ },
1870
+ {
1871
+ "dataPath": "params_shard_98.bin",
1872
+ "format": "raw-shard",
1873
+ "nbytes": 31457280,
1874
+ "records": [
1875
+ {
1876
+ "name": "model.layers.24.self_attn.qkv_proj.weight",
1877
+ "shape": [
1878
+ 5120,
1879
+ 3072
1880
+ ],
1881
+ "dtype": "bfloat16",
1882
+ "format": "raw",
1883
+ "nbytes": 31457280,
1884
+ "byteOffset": 0
1885
+ }
1886
+ ],
1887
+ "md5sum": "3aed6dd9398fa605f8c3539853ba96fa"
1888
+ },
1889
+ {
1890
+ "dataPath": "params_shard_99.bin",
1891
+ "format": "raw-shard",
1892
+ "nbytes": 18874368,
1893
+ "records": [
1894
+ {
1895
+ "name": "model.layers.24.self_attn.o_proj.weight",
1896
+ "shape": [
1897
+ 3072,
1898
+ 3072
1899
+ ],
1900
+ "dtype": "bfloat16",
1901
+ "format": "raw",
1902
+ "nbytes": 18874368,
1903
+ "byteOffset": 0
1904
+ }
1905
+ ],
1906
+ "md5sum": "6e3573ab844d10a115a672e5eb4101a1"
1907
+ },
1908
+ {
1909
+ "dataPath": "params_shard_100.bin",
1910
+ "format": "raw-shard",
1911
+ "nbytes": 50331648,
1912
+ "records": [
1913
+ {
1914
+ "name": "model.layers.25.mlp.down_proj.weight",
1915
+ "shape": [
1916
+ 3072,
1917
+ 8192
1918
+ ],
1919
+ "dtype": "bfloat16",
1920
+ "format": "raw",
1921
+ "nbytes": 50331648,
1922
+ "byteOffset": 0
1923
+ }
1924
+ ],
1925
+ "md5sum": "1f0c9184fc17fa72a166b741fbd14304"
1926
+ },
1927
+ {
1928
+ "dataPath": "params_shard_101.bin",
1929
+ "format": "raw-shard",
1930
+ "nbytes": 100663296,
1931
+ "records": [
1932
+ {
1933
+ "name": "model.layers.25.mlp.gate_up_proj.weight",
1934
+ "shape": [
1935
+ 16384,
1936
+ 3072
1937
+ ],
1938
+ "dtype": "bfloat16",
1939
+ "format": "raw",
1940
+ "nbytes": 100663296,
1941
+ "byteOffset": 0
1942
+ }
1943
+ ],
1944
+ "md5sum": "4e952489e44feff3f51f293243193963"
1945
+ },
1946
+ {
1947
+ "dataPath": "params_shard_102.bin",
1948
+ "format": "raw-shard",
1949
+ "nbytes": 31457280,
1950
+ "records": [
1951
+ {
1952
+ "name": "model.layers.25.self_attn.qkv_proj.weight",
1953
+ "shape": [
1954
+ 5120,
1955
+ 3072
1956
+ ],
1957
+ "dtype": "bfloat16",
1958
+ "format": "raw",
1959
+ "nbytes": 31457280,
1960
+ "byteOffset": 0
1961
+ }
1962
+ ],
1963
+ "md5sum": "ac0f2806df1ac7868d2e83026271f057"
1964
+ },
1965
+ {
1966
+ "dataPath": "params_shard_103.bin",
1967
+ "format": "raw-shard",
1968
+ "nbytes": 18874368,
1969
+ "records": [
1970
+ {
1971
+ "name": "model.layers.25.self_attn.o_proj.weight",
1972
+ "shape": [
1973
+ 3072,
1974
+ 3072
1975
+ ],
1976
+ "dtype": "bfloat16",
1977
+ "format": "raw",
1978
+ "nbytes": 18874368,
1979
+ "byteOffset": 0
1980
+ }
1981
+ ],
1982
+ "md5sum": "81c614dbc2fac010a3809d3c51055227"
1983
+ },
1984
+ {
1985
+ "dataPath": "params_shard_104.bin",
1986
+ "format": "raw-shard",
1987
+ "nbytes": 50331648,
1988
+ "records": [
1989
+ {
1990
+ "name": "model.layers.26.mlp.down_proj.weight",
1991
+ "shape": [
1992
+ 3072,
1993
+ 8192
1994
+ ],
1995
+ "dtype": "bfloat16",
1996
+ "format": "raw",
1997
+ "nbytes": 50331648,
1998
+ "byteOffset": 0
1999
+ }
2000
+ ],
2001
+ "md5sum": "028de1825559128df8e5fd233404f13c"
2002
+ },
2003
+ {
2004
+ "dataPath": "params_shard_105.bin",
2005
+ "format": "raw-shard",
2006
+ "nbytes": 100663296,
2007
+ "records": [
2008
+ {
2009
+ "name": "model.layers.26.mlp.gate_up_proj.weight",
2010
+ "shape": [
2011
+ 16384,
2012
+ 3072
2013
+ ],
2014
+ "dtype": "bfloat16",
2015
+ "format": "raw",
2016
+ "nbytes": 100663296,
2017
+ "byteOffset": 0
2018
+ }
2019
+ ],
2020
+ "md5sum": "91c6cc596728b4d45f07bf90e4ce40dc"
2021
+ },
2022
+ {
2023
+ "dataPath": "params_shard_106.bin",
2024
+ "format": "raw-shard",
2025
+ "nbytes": 31457280,
2026
+ "records": [
2027
+ {
2028
+ "name": "model.layers.26.self_attn.qkv_proj.weight",
2029
+ "shape": [
2030
+ 5120,
2031
+ 3072
2032
+ ],
2033
+ "dtype": "bfloat16",
2034
+ "format": "raw",
2035
+ "nbytes": 31457280,
2036
+ "byteOffset": 0
2037
+ }
2038
+ ],
2039
+ "md5sum": "ef94f2bf8dd5bcbca41d8c1cdf0a9208"
2040
+ },
2041
+ {
2042
+ "dataPath": "params_shard_107.bin",
2043
+ "format": "raw-shard",
2044
+ "nbytes": 18874368,
2045
+ "records": [
2046
+ {
2047
+ "name": "model.layers.26.self_attn.o_proj.weight",
2048
+ "shape": [
2049
+ 3072,
2050
+ 3072
2051
+ ],
2052
+ "dtype": "bfloat16",
2053
+ "format": "raw",
2054
+ "nbytes": 18874368,
2055
+ "byteOffset": 0
2056
+ }
2057
+ ],
2058
+ "md5sum": "403dbf175a8c9eec0e8f94d700500be8"
2059
+ },
2060
+ {
2061
+ "dataPath": "params_shard_108.bin",
2062
+ "format": "raw-shard",
2063
+ "nbytes": 50331648,
2064
+ "records": [
2065
+ {
2066
+ "name": "model.layers.27.mlp.down_proj.weight",
2067
+ "shape": [
2068
+ 3072,
2069
+ 8192
2070
+ ],
2071
+ "dtype": "bfloat16",
2072
+ "format": "raw",
2073
+ "nbytes": 50331648,
2074
+ "byteOffset": 0
2075
+ }
2076
+ ],
2077
+ "md5sum": "d7db1850f8f194d48ce5fca0cb8aaf7d"
2078
+ },
2079
+ {
2080
+ "dataPath": "params_shard_109.bin",
2081
+ "format": "raw-shard",
2082
+ "nbytes": 100663296,
2083
+ "records": [
2084
+ {
2085
+ "name": "model.layers.27.mlp.gate_up_proj.weight",
2086
+ "shape": [
2087
+ 16384,
2088
+ 3072
2089
+ ],
2090
+ "dtype": "bfloat16",
2091
+ "format": "raw",
2092
+ "nbytes": 100663296,
2093
+ "byteOffset": 0
2094
+ }
2095
+ ],
2096
+ "md5sum": "d970e24d360046aada8fa036e74a0104"
2097
+ },
2098
+ {
2099
+ "dataPath": "params_shard_110.bin",
2100
+ "format": "raw-shard",
2101
+ "nbytes": 31457280,
2102
+ "records": [
2103
+ {
2104
+ "name": "model.layers.27.self_attn.qkv_proj.weight",
2105
+ "shape": [
2106
+ 5120,
2107
+ 3072
2108
+ ],
2109
+ "dtype": "bfloat16",
2110
+ "format": "raw",
2111
+ "nbytes": 31457280,
2112
+ "byteOffset": 0
2113
+ }
2114
+ ],
2115
+ "md5sum": "71d768ddcd341542e6b40db9ae197d46"
2116
+ },
2117
+ {
2118
+ "dataPath": "params_shard_111.bin",
2119
+ "format": "raw-shard",
2120
+ "nbytes": 18874368,
2121
+ "records": [
2122
+ {
2123
+ "name": "model.layers.27.self_attn.o_proj.weight",
2124
+ "shape": [
2125
+ 3072,
2126
+ 3072
2127
+ ],
2128
+ "dtype": "bfloat16",
2129
+ "format": "raw",
2130
+ "nbytes": 18874368,
2131
+ "byteOffset": 0
2132
+ }
2133
+ ],
2134
+ "md5sum": "bf42c4b4af143df9f6def1800dd32225"
2135
+ },
2136
+ {
2137
+ "dataPath": "params_shard_112.bin",
2138
+ "format": "raw-shard",
2139
+ "nbytes": 31807488,
2140
+ "records": [
2141
+ {
2142
+ "name": "model.layers.0.input_layernorm.weight",
2143
+ "shape": [
2144
+ 3072
2145
+ ],
2146
+ "dtype": "bfloat16",
2147
+ "format": "raw",
2148
+ "nbytes": 6144,
2149
+ "byteOffset": 0
2150
+ },
2151
+ {
2152
+ "name": "model.layers.0.post_attention_layernorm.weight",
2153
+ "shape": [
2154
+ 3072
2155
+ ],
2156
+ "dtype": "bfloat16",
2157
+ "format": "raw",
2158
+ "nbytes": 6144,
2159
+ "byteOffset": 6144
2160
+ },
2161
+ {
2162
+ "name": "model.layers.0.self_attn.qkv_proj.weight",
2163
+ "shape": [
2164
+ 5120,
2165
+ 3072
2166
+ ],
2167
+ "dtype": "bfloat16",
2168
+ "format": "raw",
2169
+ "nbytes": 31457280,
2170
+ "byteOffset": 12288
2171
+ },
2172
+ {
2173
+ "name": "model.layers.1.input_layernorm.weight",
2174
+ "shape": [
2175
+ 3072
2176
+ ],
2177
+ "dtype": "bfloat16",
2178
+ "format": "raw",
2179
+ "nbytes": 6144,
2180
+ "byteOffset": 31469568
2181
+ },
2182
+ {
2183
+ "name": "model.layers.1.post_attention_layernorm.weight",
2184
+ "shape": [
2185
+ 3072
2186
+ ],
2187
+ "dtype": "bfloat16",
2188
+ "format": "raw",
2189
+ "nbytes": 6144,
2190
+ "byteOffset": 31475712
2191
+ },
2192
+ {
2193
+ "name": "model.layers.10.input_layernorm.weight",
2194
+ "shape": [
2195
+ 3072
2196
+ ],
2197
+ "dtype": "bfloat16",
2198
+ "format": "raw",
2199
+ "nbytes": 6144,
2200
+ "byteOffset": 31481856
2201
+ },
2202
+ {
2203
+ "name": "model.layers.10.post_attention_layernorm.weight",
2204
+ "shape": [
2205
+ 3072
2206
+ ],
2207
+ "dtype": "bfloat16",
2208
+ "format": "raw",
2209
+ "nbytes": 6144,
2210
+ "byteOffset": 31488000
2211
+ },
2212
+ {
2213
+ "name": "model.layers.11.input_layernorm.weight",
2214
+ "shape": [
2215
+ 3072
2216
+ ],
2217
+ "dtype": "bfloat16",
2218
+ "format": "raw",
2219
+ "nbytes": 6144,
2220
+ "byteOffset": 31494144
2221
+ },
2222
+ {
2223
+ "name": "model.layers.11.post_attention_layernorm.weight",
2224
+ "shape": [
2225
+ 3072
2226
+ ],
2227
+ "dtype": "bfloat16",
2228
+ "format": "raw",
2229
+ "nbytes": 6144,
2230
+ "byteOffset": 31500288
2231
+ },
2232
+ {
2233
+ "name": "model.layers.12.input_layernorm.weight",
2234
+ "shape": [
2235
+ 3072
2236
+ ],
2237
+ "dtype": "bfloat16",
2238
+ "format": "raw",
2239
+ "nbytes": 6144,
2240
+ "byteOffset": 31506432
2241
+ },
2242
+ {
2243
+ "name": "model.layers.12.post_attention_layernorm.weight",
2244
+ "shape": [
2245
+ 3072
2246
+ ],
2247
+ "dtype": "bfloat16",
2248
+ "format": "raw",
2249
+ "nbytes": 6144,
2250
+ "byteOffset": 31512576
2251
+ },
2252
+ {
2253
+ "name": "model.layers.13.input_layernorm.weight",
2254
+ "shape": [
2255
+ 3072
2256
+ ],
2257
+ "dtype": "bfloat16",
2258
+ "format": "raw",
2259
+ "nbytes": 6144,
2260
+ "byteOffset": 31518720
2261
+ },
2262
+ {
2263
+ "name": "model.layers.13.post_attention_layernorm.weight",
2264
+ "shape": [
2265
+ 3072
2266
+ ],
2267
+ "dtype": "bfloat16",
2268
+ "format": "raw",
2269
+ "nbytes": 6144,
2270
+ "byteOffset": 31524864
2271
+ },
2272
+ {
2273
+ "name": "model.layers.14.input_layernorm.weight",
2274
+ "shape": [
2275
+ 3072
2276
+ ],
2277
+ "dtype": "bfloat16",
2278
+ "format": "raw",
2279
+ "nbytes": 6144,
2280
+ "byteOffset": 31531008
2281
+ },
2282
+ {
2283
+ "name": "model.layers.14.post_attention_layernorm.weight",
2284
+ "shape": [
2285
+ 3072
2286
+ ],
2287
+ "dtype": "bfloat16",
2288
+ "format": "raw",
2289
+ "nbytes": 6144,
2290
+ "byteOffset": 31537152
2291
+ },
2292
+ {
2293
+ "name": "model.layers.15.input_layernorm.weight",
2294
+ "shape": [
2295
+ 3072
2296
+ ],
2297
+ "dtype": "bfloat16",
2298
+ "format": "raw",
2299
+ "nbytes": 6144,
2300
+ "byteOffset": 31543296
2301
+ },
2302
+ {
2303
+ "name": "model.layers.15.post_attention_layernorm.weight",
2304
+ "shape": [
2305
+ 3072
2306
+ ],
2307
+ "dtype": "bfloat16",
2308
+ "format": "raw",
2309
+ "nbytes": 6144,
2310
+ "byteOffset": 31549440
2311
+ },
2312
+ {
2313
+ "name": "model.layers.16.input_layernorm.weight",
2314
+ "shape": [
2315
+ 3072
2316
+ ],
2317
+ "dtype": "bfloat16",
2318
+ "format": "raw",
2319
+ "nbytes": 6144,
2320
+ "byteOffset": 31555584
2321
+ },
2322
+ {
2323
+ "name": "model.layers.16.post_attention_layernorm.weight",
2324
+ "shape": [
2325
+ 3072
2326
+ ],
2327
+ "dtype": "bfloat16",
2328
+ "format": "raw",
2329
+ "nbytes": 6144,
2330
+ "byteOffset": 31561728
2331
+ },
2332
+ {
2333
+ "name": "model.layers.17.input_layernorm.weight",
2334
+ "shape": [
2335
+ 3072
2336
+ ],
2337
+ "dtype": "bfloat16",
2338
+ "format": "raw",
2339
+ "nbytes": 6144,
2340
+ "byteOffset": 31567872
2341
+ },
2342
+ {
2343
+ "name": "model.layers.17.post_attention_layernorm.weight",
2344
+ "shape": [
2345
+ 3072
2346
+ ],
2347
+ "dtype": "bfloat16",
2348
+ "format": "raw",
2349
+ "nbytes": 6144,
2350
+ "byteOffset": 31574016
2351
+ },
2352
+ {
2353
+ "name": "model.layers.18.input_layernorm.weight",
2354
+ "shape": [
2355
+ 3072
2356
+ ],
2357
+ "dtype": "bfloat16",
2358
+ "format": "raw",
2359
+ "nbytes": 6144,
2360
+ "byteOffset": 31580160
2361
+ },
2362
+ {
2363
+ "name": "model.layers.18.post_attention_layernorm.weight",
2364
+ "shape": [
2365
+ 3072
2366
+ ],
2367
+ "dtype": "bfloat16",
2368
+ "format": "raw",
2369
+ "nbytes": 6144,
2370
+ "byteOffset": 31586304
2371
+ },
2372
+ {
2373
+ "name": "model.layers.19.input_layernorm.weight",
2374
+ "shape": [
2375
+ 3072
2376
+ ],
2377
+ "dtype": "bfloat16",
2378
+ "format": "raw",
2379
+ "nbytes": 6144,
2380
+ "byteOffset": 31592448
2381
+ },
2382
+ {
2383
+ "name": "model.layers.19.post_attention_layernorm.weight",
2384
+ "shape": [
2385
+ 3072
2386
+ ],
2387
+ "dtype": "bfloat16",
2388
+ "format": "raw",
2389
+ "nbytes": 6144,
2390
+ "byteOffset": 31598592
2391
+ },
2392
+ {
2393
+ "name": "model.layers.2.input_layernorm.weight",
2394
+ "shape": [
2395
+ 3072
2396
+ ],
2397
+ "dtype": "bfloat16",
2398
+ "format": "raw",
2399
+ "nbytes": 6144,
2400
+ "byteOffset": 31604736
2401
+ },
2402
+ {
2403
+ "name": "model.layers.2.post_attention_layernorm.weight",
2404
+ "shape": [
2405
+ 3072
2406
+ ],
2407
+ "dtype": "bfloat16",
2408
+ "format": "raw",
2409
+ "nbytes": 6144,
2410
+ "byteOffset": 31610880
2411
+ },
2412
+ {
2413
+ "name": "model.layers.3.input_layernorm.weight",
2414
+ "shape": [
2415
+ 3072
2416
+ ],
2417
+ "dtype": "bfloat16",
2418
+ "format": "raw",
2419
+ "nbytes": 6144,
2420
+ "byteOffset": 31617024
2421
+ },
2422
+ {
2423
+ "name": "model.layers.3.post_attention_layernorm.weight",
2424
+ "shape": [
2425
+ 3072
2426
+ ],
2427
+ "dtype": "bfloat16",
2428
+ "format": "raw",
2429
+ "nbytes": 6144,
2430
+ "byteOffset": 31623168
2431
+ },
2432
+ {
2433
+ "name": "model.layers.4.input_layernorm.weight",
2434
+ "shape": [
2435
+ 3072
2436
+ ],
2437
+ "dtype": "bfloat16",
2438
+ "format": "raw",
2439
+ "nbytes": 6144,
2440
+ "byteOffset": 31629312
2441
+ },
2442
+ {
2443
+ "name": "model.layers.4.post_attention_layernorm.weight",
2444
+ "shape": [
2445
+ 3072
2446
+ ],
2447
+ "dtype": "bfloat16",
2448
+ "format": "raw",
2449
+ "nbytes": 6144,
2450
+ "byteOffset": 31635456
2451
+ },
2452
+ {
2453
+ "name": "model.layers.5.input_layernorm.weight",
2454
+ "shape": [
2455
+ 3072
2456
+ ],
2457
+ "dtype": "bfloat16",
2458
+ "format": "raw",
2459
+ "nbytes": 6144,
2460
+ "byteOffset": 31641600
2461
+ },
2462
+ {
2463
+ "name": "model.layers.5.post_attention_layernorm.weight",
2464
+ "shape": [
2465
+ 3072
2466
+ ],
2467
+ "dtype": "bfloat16",
2468
+ "format": "raw",
2469
+ "nbytes": 6144,
2470
+ "byteOffset": 31647744
2471
+ },
2472
+ {
2473
+ "name": "model.layers.6.input_layernorm.weight",
2474
+ "shape": [
2475
+ 3072
2476
+ ],
2477
+ "dtype": "bfloat16",
2478
+ "format": "raw",
2479
+ "nbytes": 6144,
2480
+ "byteOffset": 31653888
2481
+ },
2482
+ {
2483
+ "name": "model.layers.6.post_attention_layernorm.weight",
2484
+ "shape": [
2485
+ 3072
2486
+ ],
2487
+ "dtype": "bfloat16",
2488
+ "format": "raw",
2489
+ "nbytes": 6144,
2490
+ "byteOffset": 31660032
2491
+ },
2492
+ {
2493
+ "name": "model.layers.7.input_layernorm.weight",
2494
+ "shape": [
2495
+ 3072
2496
+ ],
2497
+ "dtype": "bfloat16",
2498
+ "format": "raw",
2499
+ "nbytes": 6144,
2500
+ "byteOffset": 31666176
2501
+ },
2502
+ {
2503
+ "name": "model.layers.7.post_attention_layernorm.weight",
2504
+ "shape": [
2505
+ 3072
2506
+ ],
2507
+ "dtype": "bfloat16",
2508
+ "format": "raw",
2509
+ "nbytes": 6144,
2510
+ "byteOffset": 31672320
2511
+ },
2512
+ {
2513
+ "name": "model.layers.8.input_layernorm.weight",
2514
+ "shape": [
2515
+ 3072
2516
+ ],
2517
+ "dtype": "bfloat16",
2518
+ "format": "raw",
2519
+ "nbytes": 6144,
2520
+ "byteOffset": 31678464
2521
+ },
2522
+ {
2523
+ "name": "model.layers.8.post_attention_layernorm.weight",
2524
+ "shape": [
2525
+ 3072
2526
+ ],
2527
+ "dtype": "bfloat16",
2528
+ "format": "raw",
2529
+ "nbytes": 6144,
2530
+ "byteOffset": 31684608
2531
+ },
2532
+ {
2533
+ "name": "model.layers.9.input_layernorm.weight",
2534
+ "shape": [
2535
+ 3072
2536
+ ],
2537
+ "dtype": "bfloat16",
2538
+ "format": "raw",
2539
+ "nbytes": 6144,
2540
+ "byteOffset": 31690752
2541
+ },
2542
+ {
2543
+ "name": "model.layers.9.post_attention_layernorm.weight",
2544
+ "shape": [
2545
+ 3072
2546
+ ],
2547
+ "dtype": "bfloat16",
2548
+ "format": "raw",
2549
+ "nbytes": 6144,
2550
+ "byteOffset": 31696896
2551
+ },
2552
+ {
2553
+ "name": "model.layers.20.input_layernorm.weight",
2554
+ "shape": [
2555
+ 3072
2556
+ ],
2557
+ "dtype": "bfloat16",
2558
+ "format": "raw",
2559
+ "nbytes": 6144,
2560
+ "byteOffset": 31703040
2561
+ },
2562
+ {
2563
+ "name": "model.layers.20.post_attention_layernorm.weight",
2564
+ "shape": [
2565
+ 3072
2566
+ ],
2567
+ "dtype": "bfloat16",
2568
+ "format": "raw",
2569
+ "nbytes": 6144,
2570
+ "byteOffset": 31709184
2571
+ },
2572
+ {
2573
+ "name": "model.layers.21.input_layernorm.weight",
2574
+ "shape": [
2575
+ 3072
2576
+ ],
2577
+ "dtype": "bfloat16",
2578
+ "format": "raw",
2579
+ "nbytes": 6144,
2580
+ "byteOffset": 31715328
2581
+ },
2582
+ {
2583
+ "name": "model.layers.21.post_attention_layernorm.weight",
2584
+ "shape": [
2585
+ 3072
2586
+ ],
2587
+ "dtype": "bfloat16",
2588
+ "format": "raw",
2589
+ "nbytes": 6144,
2590
+ "byteOffset": 31721472
2591
+ },
2592
+ {
2593
+ "name": "model.layers.22.input_layernorm.weight",
2594
+ "shape": [
2595
+ 3072
2596
+ ],
2597
+ "dtype": "bfloat16",
2598
+ "format": "raw",
2599
+ "nbytes": 6144,
2600
+ "byteOffset": 31727616
2601
+ },
2602
+ {
2603
+ "name": "model.layers.22.post_attention_layernorm.weight",
2604
+ "shape": [
2605
+ 3072
2606
+ ],
2607
+ "dtype": "bfloat16",
2608
+ "format": "raw",
2609
+ "nbytes": 6144,
2610
+ "byteOffset": 31733760
2611
+ },
2612
+ {
2613
+ "name": "model.layers.23.input_layernorm.weight",
2614
+ "shape": [
2615
+ 3072
2616
+ ],
2617
+ "dtype": "bfloat16",
2618
+ "format": "raw",
2619
+ "nbytes": 6144,
2620
+ "byteOffset": 31739904
2621
+ },
2622
+ {
2623
+ "name": "model.layers.23.post_attention_layernorm.weight",
2624
+ "shape": [
2625
+ 3072
2626
+ ],
2627
+ "dtype": "bfloat16",
2628
+ "format": "raw",
2629
+ "nbytes": 6144,
2630
+ "byteOffset": 31746048
2631
+ },
2632
+ {
2633
+ "name": "model.layers.24.input_layernorm.weight",
2634
+ "shape": [
2635
+ 3072
2636
+ ],
2637
+ "dtype": "bfloat16",
2638
+ "format": "raw",
2639
+ "nbytes": 6144,
2640
+ "byteOffset": 31752192
2641
+ },
2642
+ {
2643
+ "name": "model.layers.24.post_attention_layernorm.weight",
2644
+ "shape": [
2645
+ 3072
2646
+ ],
2647
+ "dtype": "bfloat16",
2648
+ "format": "raw",
2649
+ "nbytes": 6144,
2650
+ "byteOffset": 31758336
2651
+ },
2652
+ {
2653
+ "name": "model.layers.25.input_layernorm.weight",
2654
+ "shape": [
2655
+ 3072
2656
+ ],
2657
+ "dtype": "bfloat16",
2658
+ "format": "raw",
2659
+ "nbytes": 6144,
2660
+ "byteOffset": 31764480
2661
+ },
2662
+ {
2663
+ "name": "model.layers.25.post_attention_layernorm.weight",
2664
+ "shape": [
2665
+ 3072
2666
+ ],
2667
+ "dtype": "bfloat16",
2668
+ "format": "raw",
2669
+ "nbytes": 6144,
2670
+ "byteOffset": 31770624
2671
+ },
2672
+ {
2673
+ "name": "model.layers.26.input_layernorm.weight",
2674
+ "shape": [
2675
+ 3072
2676
+ ],
2677
+ "dtype": "bfloat16",
2678
+ "format": "raw",
2679
+ "nbytes": 6144,
2680
+ "byteOffset": 31776768
2681
+ },
2682
+ {
2683
+ "name": "model.layers.26.post_attention_layernorm.weight",
2684
+ "shape": [
2685
+ 3072
2686
+ ],
2687
+ "dtype": "bfloat16",
2688
+ "format": "raw",
2689
+ "nbytes": 6144,
2690
+ "byteOffset": 31782912
2691
+ },
2692
+ {
2693
+ "name": "model.layers.27.input_layernorm.weight",
2694
+ "shape": [
2695
+ 3072
2696
+ ],
2697
+ "dtype": "bfloat16",
2698
+ "format": "raw",
2699
+ "nbytes": 6144,
2700
+ "byteOffset": 31789056
2701
+ },
2702
+ {
2703
+ "name": "model.layers.27.post_attention_layernorm.weight",
2704
+ "shape": [
2705
+ 3072
2706
+ ],
2707
+ "dtype": "bfloat16",
2708
+ "format": "raw",
2709
+ "nbytes": 6144,
2710
+ "byteOffset": 31795200
2711
+ },
2712
+ {
2713
+ "name": "model.norm.weight",
2714
+ "shape": [
2715
+ 3072
2716
+ ],
2717
+ "dtype": "bfloat16",
2718
+ "format": "raw",
2719
+ "nbytes": 6144,
2720
+ "byteOffset": 31801344
2721
+ }
2722
+ ],
2723
+ "md5sum": "8952ab985c888d6241e0e6204be82d7a"
2724
+ }
2725
+ ]
2726
+ }
ndarray-cache.json ADDED
@@ -0,0 +1,2726 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "ParamSize": 170,
4
+ "ParamBytes": 12850999296.0,
5
+ "BitsPerParam": 32.0
6
+ },
7
+ "records": [
8
+ {
9
+ "dataPath": "params_shard_0.bin",
10
+ "format": "raw-shard",
11
+ "nbytes": 788004864,
12
+ "records": [
13
+ {
14
+ "name": "model.embed_tokens.weight",
15
+ "shape": [
16
+ 128256,
17
+ 3072
18
+ ],
19
+ "dtype": "float32",
20
+ "format": "f32-to-bf16",
21
+ "nbytes": 788004864,
22
+ "byteOffset": 0
23
+ }
24
+ ],
25
+ "md5sum": "133c0866b77e2e5678be91a753bb7e28"
26
+ },
27
+ {
28
+ "dataPath": "params_shard_1.bin",
29
+ "format": "raw-shard",
30
+ "nbytes": 50331648,
31
+ "records": [
32
+ {
33
+ "name": "model.layers.0.mlp.down_proj.weight",
34
+ "shape": [
35
+ 3072,
36
+ 8192
37
+ ],
38
+ "dtype": "float32",
39
+ "format": "f32-to-bf16",
40
+ "nbytes": 50331648,
41
+ "byteOffset": 0
42
+ }
43
+ ],
44
+ "md5sum": "93080db9b85b8252e1138672b9902f9a"
45
+ },
46
+ {
47
+ "dataPath": "params_shard_2.bin",
48
+ "format": "raw-shard",
49
+ "nbytes": 100663296,
50
+ "records": [
51
+ {
52
+ "name": "model.layers.0.mlp.gate_up_proj.weight",
53
+ "shape": [
54
+ 16384,
55
+ 3072
56
+ ],
57
+ "dtype": "float32",
58
+ "format": "f32-to-bf16",
59
+ "nbytes": 100663296,
60
+ "byteOffset": 0
61
+ }
62
+ ],
63
+ "md5sum": "0c18f71e770e7c7c07270f2d7ec06d1c"
64
+ },
65
+ {
66
+ "dataPath": "params_shard_3.bin",
67
+ "format": "raw-shard",
68
+ "nbytes": 18874368,
69
+ "records": [
70
+ {
71
+ "name": "model.layers.0.self_attn.o_proj.weight",
72
+ "shape": [
73
+ 3072,
74
+ 3072
75
+ ],
76
+ "dtype": "float32",
77
+ "format": "f32-to-bf16",
78
+ "nbytes": 18874368,
79
+ "byteOffset": 0
80
+ }
81
+ ],
82
+ "md5sum": "44251b37a994062bedadfcd599328de2"
83
+ },
84
+ {
85
+ "dataPath": "params_shard_4.bin",
86
+ "format": "raw-shard",
87
+ "nbytes": 50331648,
88
+ "records": [
89
+ {
90
+ "name": "model.layers.1.mlp.down_proj.weight",
91
+ "shape": [
92
+ 3072,
93
+ 8192
94
+ ],
95
+ "dtype": "float32",
96
+ "format": "f32-to-bf16",
97
+ "nbytes": 50331648,
98
+ "byteOffset": 0
99
+ }
100
+ ],
101
+ "md5sum": "bf00d477f6d478010a9c7584c93723fd"
102
+ },
103
+ {
104
+ "dataPath": "params_shard_5.bin",
105
+ "format": "raw-shard",
106
+ "nbytes": 100663296,
107
+ "records": [
108
+ {
109
+ "name": "model.layers.1.mlp.gate_up_proj.weight",
110
+ "shape": [
111
+ 16384,
112
+ 3072
113
+ ],
114
+ "dtype": "float32",
115
+ "format": "f32-to-bf16",
116
+ "nbytes": 100663296,
117
+ "byteOffset": 0
118
+ }
119
+ ],
120
+ "md5sum": "23a00478c5789710e44b2229f5b1f0ea"
121
+ },
122
+ {
123
+ "dataPath": "params_shard_6.bin",
124
+ "format": "raw-shard",
125
+ "nbytes": 31457280,
126
+ "records": [
127
+ {
128
+ "name": "model.layers.1.self_attn.qkv_proj.weight",
129
+ "shape": [
130
+ 5120,
131
+ 3072
132
+ ],
133
+ "dtype": "float32",
134
+ "format": "f32-to-bf16",
135
+ "nbytes": 31457280,
136
+ "byteOffset": 0
137
+ }
138
+ ],
139
+ "md5sum": "c15617b1ce4de0ecc8e85d8566d4fd07"
140
+ },
141
+ {
142
+ "dataPath": "params_shard_7.bin",
143
+ "format": "raw-shard",
144
+ "nbytes": 18874368,
145
+ "records": [
146
+ {
147
+ "name": "model.layers.1.self_attn.o_proj.weight",
148
+ "shape": [
149
+ 3072,
150
+ 3072
151
+ ],
152
+ "dtype": "float32",
153
+ "format": "f32-to-bf16",
154
+ "nbytes": 18874368,
155
+ "byteOffset": 0
156
+ }
157
+ ],
158
+ "md5sum": "7926239d43547e6b686b15951a3a84e7"
159
+ },
160
+ {
161
+ "dataPath": "params_shard_8.bin",
162
+ "format": "raw-shard",
163
+ "nbytes": 50331648,
164
+ "records": [
165
+ {
166
+ "name": "model.layers.10.mlp.down_proj.weight",
167
+ "shape": [
168
+ 3072,
169
+ 8192
170
+ ],
171
+ "dtype": "float32",
172
+ "format": "f32-to-bf16",
173
+ "nbytes": 50331648,
174
+ "byteOffset": 0
175
+ }
176
+ ],
177
+ "md5sum": "2237119d8791ce2a58941361b80fa84d"
178
+ },
179
+ {
180
+ "dataPath": "params_shard_9.bin",
181
+ "format": "raw-shard",
182
+ "nbytes": 100663296,
183
+ "records": [
184
+ {
185
+ "name": "model.layers.10.mlp.gate_up_proj.weight",
186
+ "shape": [
187
+ 16384,
188
+ 3072
189
+ ],
190
+ "dtype": "float32",
191
+ "format": "f32-to-bf16",
192
+ "nbytes": 100663296,
193
+ "byteOffset": 0
194
+ }
195
+ ],
196
+ "md5sum": "d70ec745e1feafbb4a1d86914adfca38"
197
+ },
198
+ {
199
+ "dataPath": "params_shard_10.bin",
200
+ "format": "raw-shard",
201
+ "nbytes": 31457280,
202
+ "records": [
203
+ {
204
+ "name": "model.layers.10.self_attn.qkv_proj.weight",
205
+ "shape": [
206
+ 5120,
207
+ 3072
208
+ ],
209
+ "dtype": "float32",
210
+ "format": "f32-to-bf16",
211
+ "nbytes": 31457280,
212
+ "byteOffset": 0
213
+ }
214
+ ],
215
+ "md5sum": "75f3e0d41780aa8602f46b1016314c05"
216
+ },
217
+ {
218
+ "dataPath": "params_shard_11.bin",
219
+ "format": "raw-shard",
220
+ "nbytes": 18874368,
221
+ "records": [
222
+ {
223
+ "name": "model.layers.10.self_attn.o_proj.weight",
224
+ "shape": [
225
+ 3072,
226
+ 3072
227
+ ],
228
+ "dtype": "float32",
229
+ "format": "f32-to-bf16",
230
+ "nbytes": 18874368,
231
+ "byteOffset": 0
232
+ }
233
+ ],
234
+ "md5sum": "90f3ce08908231682aa92f2c1296e5fc"
235
+ },
236
+ {
237
+ "dataPath": "params_shard_12.bin",
238
+ "format": "raw-shard",
239
+ "nbytes": 50331648,
240
+ "records": [
241
+ {
242
+ "name": "model.layers.11.mlp.down_proj.weight",
243
+ "shape": [
244
+ 3072,
245
+ 8192
246
+ ],
247
+ "dtype": "float32",
248
+ "format": "f32-to-bf16",
249
+ "nbytes": 50331648,
250
+ "byteOffset": 0
251
+ }
252
+ ],
253
+ "md5sum": "8e2ea40c1f92cbb61b9d658895140b07"
254
+ },
255
+ {
256
+ "dataPath": "params_shard_13.bin",
257
+ "format": "raw-shard",
258
+ "nbytes": 100663296,
259
+ "records": [
260
+ {
261
+ "name": "model.layers.11.mlp.gate_up_proj.weight",
262
+ "shape": [
263
+ 16384,
264
+ 3072
265
+ ],
266
+ "dtype": "float32",
267
+ "format": "f32-to-bf16",
268
+ "nbytes": 100663296,
269
+ "byteOffset": 0
270
+ }
271
+ ],
272
+ "md5sum": "083ec2e27ba602a63ebb32ce85aff2db"
273
+ },
274
+ {
275
+ "dataPath": "params_shard_14.bin",
276
+ "format": "raw-shard",
277
+ "nbytes": 31457280,
278
+ "records": [
279
+ {
280
+ "name": "model.layers.11.self_attn.qkv_proj.weight",
281
+ "shape": [
282
+ 5120,
283
+ 3072
284
+ ],
285
+ "dtype": "float32",
286
+ "format": "f32-to-bf16",
287
+ "nbytes": 31457280,
288
+ "byteOffset": 0
289
+ }
290
+ ],
291
+ "md5sum": "94dd82b511ec123317e8ece8285765af"
292
+ },
293
+ {
294
+ "dataPath": "params_shard_15.bin",
295
+ "format": "raw-shard",
296
+ "nbytes": 18874368,
297
+ "records": [
298
+ {
299
+ "name": "model.layers.11.self_attn.o_proj.weight",
300
+ "shape": [
301
+ 3072,
302
+ 3072
303
+ ],
304
+ "dtype": "float32",
305
+ "format": "f32-to-bf16",
306
+ "nbytes": 18874368,
307
+ "byteOffset": 0
308
+ }
309
+ ],
310
+ "md5sum": "75fc3f6b90b60616d2696b12412fc23b"
311
+ },
312
+ {
313
+ "dataPath": "params_shard_16.bin",
314
+ "format": "raw-shard",
315
+ "nbytes": 50331648,
316
+ "records": [
317
+ {
318
+ "name": "model.layers.12.mlp.down_proj.weight",
319
+ "shape": [
320
+ 3072,
321
+ 8192
322
+ ],
323
+ "dtype": "float32",
324
+ "format": "f32-to-bf16",
325
+ "nbytes": 50331648,
326
+ "byteOffset": 0
327
+ }
328
+ ],
329
+ "md5sum": "1f5e4e0701c74744048f742778298f61"
330
+ },
331
+ {
332
+ "dataPath": "params_shard_17.bin",
333
+ "format": "raw-shard",
334
+ "nbytes": 100663296,
335
+ "records": [
336
+ {
337
+ "name": "model.layers.12.mlp.gate_up_proj.weight",
338
+ "shape": [
339
+ 16384,
340
+ 3072
341
+ ],
342
+ "dtype": "float32",
343
+ "format": "f32-to-bf16",
344
+ "nbytes": 100663296,
345
+ "byteOffset": 0
346
+ }
347
+ ],
348
+ "md5sum": "91108c2df99c4b9b87f27edc7c75d600"
349
+ },
350
+ {
351
+ "dataPath": "params_shard_18.bin",
352
+ "format": "raw-shard",
353
+ "nbytes": 31457280,
354
+ "records": [
355
+ {
356
+ "name": "model.layers.12.self_attn.qkv_proj.weight",
357
+ "shape": [
358
+ 5120,
359
+ 3072
360
+ ],
361
+ "dtype": "float32",
362
+ "format": "f32-to-bf16",
363
+ "nbytes": 31457280,
364
+ "byteOffset": 0
365
+ }
366
+ ],
367
+ "md5sum": "a8abdbceeb4dd058614f3f144a77ece2"
368
+ },
369
+ {
370
+ "dataPath": "params_shard_19.bin",
371
+ "format": "raw-shard",
372
+ "nbytes": 18874368,
373
+ "records": [
374
+ {
375
+ "name": "model.layers.12.self_attn.o_proj.weight",
376
+ "shape": [
377
+ 3072,
378
+ 3072
379
+ ],
380
+ "dtype": "float32",
381
+ "format": "f32-to-bf16",
382
+ "nbytes": 18874368,
383
+ "byteOffset": 0
384
+ }
385
+ ],
386
+ "md5sum": "c75eab31f6da7c631a17e2ce852cf003"
387
+ },
388
+ {
389
+ "dataPath": "params_shard_20.bin",
390
+ "format": "raw-shard",
391
+ "nbytes": 50331648,
392
+ "records": [
393
+ {
394
+ "name": "model.layers.13.mlp.down_proj.weight",
395
+ "shape": [
396
+ 3072,
397
+ 8192
398
+ ],
399
+ "dtype": "float32",
400
+ "format": "f32-to-bf16",
401
+ "nbytes": 50331648,
402
+ "byteOffset": 0
403
+ }
404
+ ],
405
+ "md5sum": "c6dbd24e7eb2d5485a7fa009326ca523"
406
+ },
407
+ {
408
+ "dataPath": "params_shard_21.bin",
409
+ "format": "raw-shard",
410
+ "nbytes": 100663296,
411
+ "records": [
412
+ {
413
+ "name": "model.layers.13.mlp.gate_up_proj.weight",
414
+ "shape": [
415
+ 16384,
416
+ 3072
417
+ ],
418
+ "dtype": "float32",
419
+ "format": "f32-to-bf16",
420
+ "nbytes": 100663296,
421
+ "byteOffset": 0
422
+ }
423
+ ],
424
+ "md5sum": "ce2e8339c0cb6d4457a5ede2b3bee0b5"
425
+ },
426
+ {
427
+ "dataPath": "params_shard_22.bin",
428
+ "format": "raw-shard",
429
+ "nbytes": 31457280,
430
+ "records": [
431
+ {
432
+ "name": "model.layers.13.self_attn.qkv_proj.weight",
433
+ "shape": [
434
+ 5120,
435
+ 3072
436
+ ],
437
+ "dtype": "float32",
438
+ "format": "f32-to-bf16",
439
+ "nbytes": 31457280,
440
+ "byteOffset": 0
441
+ }
442
+ ],
443
+ "md5sum": "611b51469fd096b07a9463d8f2a664eb"
444
+ },
445
+ {
446
+ "dataPath": "params_shard_23.bin",
447
+ "format": "raw-shard",
448
+ "nbytes": 18874368,
449
+ "records": [
450
+ {
451
+ "name": "model.layers.13.self_attn.o_proj.weight",
452
+ "shape": [
453
+ 3072,
454
+ 3072
455
+ ],
456
+ "dtype": "float32",
457
+ "format": "f32-to-bf16",
458
+ "nbytes": 18874368,
459
+ "byteOffset": 0
460
+ }
461
+ ],
462
+ "md5sum": "50dc7e6b1fbee76eaa87ce0f5ec3fcc0"
463
+ },
464
+ {
465
+ "dataPath": "params_shard_24.bin",
466
+ "format": "raw-shard",
467
+ "nbytes": 50331648,
468
+ "records": [
469
+ {
470
+ "name": "model.layers.14.mlp.down_proj.weight",
471
+ "shape": [
472
+ 3072,
473
+ 8192
474
+ ],
475
+ "dtype": "float32",
476
+ "format": "f32-to-bf16",
477
+ "nbytes": 50331648,
478
+ "byteOffset": 0
479
+ }
480
+ ],
481
+ "md5sum": "c6f7f3faeaef9e79991b7d94b25ee011"
482
+ },
483
+ {
484
+ "dataPath": "params_shard_25.bin",
485
+ "format": "raw-shard",
486
+ "nbytes": 100663296,
487
+ "records": [
488
+ {
489
+ "name": "model.layers.14.mlp.gate_up_proj.weight",
490
+ "shape": [
491
+ 16384,
492
+ 3072
493
+ ],
494
+ "dtype": "float32",
495
+ "format": "f32-to-bf16",
496
+ "nbytes": 100663296,
497
+ "byteOffset": 0
498
+ }
499
+ ],
500
+ "md5sum": "815b14c6b06f48a87b5adc0cca956cfd"
501
+ },
502
+ {
503
+ "dataPath": "params_shard_26.bin",
504
+ "format": "raw-shard",
505
+ "nbytes": 31457280,
506
+ "records": [
507
+ {
508
+ "name": "model.layers.14.self_attn.qkv_proj.weight",
509
+ "shape": [
510
+ 5120,
511
+ 3072
512
+ ],
513
+ "dtype": "float32",
514
+ "format": "f32-to-bf16",
515
+ "nbytes": 31457280,
516
+ "byteOffset": 0
517
+ }
518
+ ],
519
+ "md5sum": "0916f5bed827fa16ce1ebe9c089097d4"
520
+ },
521
+ {
522
+ "dataPath": "params_shard_27.bin",
523
+ "format": "raw-shard",
524
+ "nbytes": 18874368,
525
+ "records": [
526
+ {
527
+ "name": "model.layers.14.self_attn.o_proj.weight",
528
+ "shape": [
529
+ 3072,
530
+ 3072
531
+ ],
532
+ "dtype": "float32",
533
+ "format": "f32-to-bf16",
534
+ "nbytes": 18874368,
535
+ "byteOffset": 0
536
+ }
537
+ ],
538
+ "md5sum": "9ddd05a218bf06c0868f80bd0b270eda"
539
+ },
540
+ {
541
+ "dataPath": "params_shard_28.bin",
542
+ "format": "raw-shard",
543
+ "nbytes": 50331648,
544
+ "records": [
545
+ {
546
+ "name": "model.layers.15.mlp.down_proj.weight",
547
+ "shape": [
548
+ 3072,
549
+ 8192
550
+ ],
551
+ "dtype": "float32",
552
+ "format": "f32-to-bf16",
553
+ "nbytes": 50331648,
554
+ "byteOffset": 0
555
+ }
556
+ ],
557
+ "md5sum": "9e02113405c9b93e9eecb883f1db14d5"
558
+ },
559
+ {
560
+ "dataPath": "params_shard_29.bin",
561
+ "format": "raw-shard",
562
+ "nbytes": 100663296,
563
+ "records": [
564
+ {
565
+ "name": "model.layers.15.mlp.gate_up_proj.weight",
566
+ "shape": [
567
+ 16384,
568
+ 3072
569
+ ],
570
+ "dtype": "float32",
571
+ "format": "f32-to-bf16",
572
+ "nbytes": 100663296,
573
+ "byteOffset": 0
574
+ }
575
+ ],
576
+ "md5sum": "303e6706d5782c66c3a11756ee513dbb"
577
+ },
578
+ {
579
+ "dataPath": "params_shard_30.bin",
580
+ "format": "raw-shard",
581
+ "nbytes": 31457280,
582
+ "records": [
583
+ {
584
+ "name": "model.layers.15.self_attn.qkv_proj.weight",
585
+ "shape": [
586
+ 5120,
587
+ 3072
588
+ ],
589
+ "dtype": "float32",
590
+ "format": "f32-to-bf16",
591
+ "nbytes": 31457280,
592
+ "byteOffset": 0
593
+ }
594
+ ],
595
+ "md5sum": "59c43454b5e413cbaab867a90ef827f3"
596
+ },
597
+ {
598
+ "dataPath": "params_shard_31.bin",
599
+ "format": "raw-shard",
600
+ "nbytes": 18874368,
601
+ "records": [
602
+ {
603
+ "name": "model.layers.15.self_attn.o_proj.weight",
604
+ "shape": [
605
+ 3072,
606
+ 3072
607
+ ],
608
+ "dtype": "float32",
609
+ "format": "f32-to-bf16",
610
+ "nbytes": 18874368,
611
+ "byteOffset": 0
612
+ }
613
+ ],
614
+ "md5sum": "f148aa81afde44c61e15ad4ddf01c8a9"
615
+ },
616
+ {
617
+ "dataPath": "params_shard_32.bin",
618
+ "format": "raw-shard",
619
+ "nbytes": 50331648,
620
+ "records": [
621
+ {
622
+ "name": "model.layers.16.mlp.down_proj.weight",
623
+ "shape": [
624
+ 3072,
625
+ 8192
626
+ ],
627
+ "dtype": "float32",
628
+ "format": "f32-to-bf16",
629
+ "nbytes": 50331648,
630
+ "byteOffset": 0
631
+ }
632
+ ],
633
+ "md5sum": "c1bc09fcd5aedc98f925795d44bd597a"
634
+ },
635
+ {
636
+ "dataPath": "params_shard_33.bin",
637
+ "format": "raw-shard",
638
+ "nbytes": 100663296,
639
+ "records": [
640
+ {
641
+ "name": "model.layers.16.mlp.gate_up_proj.weight",
642
+ "shape": [
643
+ 16384,
644
+ 3072
645
+ ],
646
+ "dtype": "float32",
647
+ "format": "f32-to-bf16",
648
+ "nbytes": 100663296,
649
+ "byteOffset": 0
650
+ }
651
+ ],
652
+ "md5sum": "d75aa45200b65bc44500abb95e9efdcc"
653
+ },
654
+ {
655
+ "dataPath": "params_shard_34.bin",
656
+ "format": "raw-shard",
657
+ "nbytes": 31457280,
658
+ "records": [
659
+ {
660
+ "name": "model.layers.16.self_attn.qkv_proj.weight",
661
+ "shape": [
662
+ 5120,
663
+ 3072
664
+ ],
665
+ "dtype": "float32",
666
+ "format": "f32-to-bf16",
667
+ "nbytes": 31457280,
668
+ "byteOffset": 0
669
+ }
670
+ ],
671
+ "md5sum": "d2607a22d2aaa52916e1f99977674d3c"
672
+ },
673
+ {
674
+ "dataPath": "params_shard_35.bin",
675
+ "format": "raw-shard",
676
+ "nbytes": 18874368,
677
+ "records": [
678
+ {
679
+ "name": "model.layers.16.self_attn.o_proj.weight",
680
+ "shape": [
681
+ 3072,
682
+ 3072
683
+ ],
684
+ "dtype": "float32",
685
+ "format": "f32-to-bf16",
686
+ "nbytes": 18874368,
687
+ "byteOffset": 0
688
+ }
689
+ ],
690
+ "md5sum": "6b0bc187e174276f9068c71f001861af"
691
+ },
692
+ {
693
+ "dataPath": "params_shard_36.bin",
694
+ "format": "raw-shard",
695
+ "nbytes": 50331648,
696
+ "records": [
697
+ {
698
+ "name": "model.layers.17.mlp.down_proj.weight",
699
+ "shape": [
700
+ 3072,
701
+ 8192
702
+ ],
703
+ "dtype": "float32",
704
+ "format": "f32-to-bf16",
705
+ "nbytes": 50331648,
706
+ "byteOffset": 0
707
+ }
708
+ ],
709
+ "md5sum": "f29349e71e3c7cc0c5172fb8de9e637f"
710
+ },
711
+ {
712
+ "dataPath": "params_shard_37.bin",
713
+ "format": "raw-shard",
714
+ "nbytes": 100663296,
715
+ "records": [
716
+ {
717
+ "name": "model.layers.17.mlp.gate_up_proj.weight",
718
+ "shape": [
719
+ 16384,
720
+ 3072
721
+ ],
722
+ "dtype": "float32",
723
+ "format": "f32-to-bf16",
724
+ "nbytes": 100663296,
725
+ "byteOffset": 0
726
+ }
727
+ ],
728
+ "md5sum": "b98a90817c9fdb702470209e341ddd5c"
729
+ },
730
+ {
731
+ "dataPath": "params_shard_38.bin",
732
+ "format": "raw-shard",
733
+ "nbytes": 31457280,
734
+ "records": [
735
+ {
736
+ "name": "model.layers.17.self_attn.qkv_proj.weight",
737
+ "shape": [
738
+ 5120,
739
+ 3072
740
+ ],
741
+ "dtype": "float32",
742
+ "format": "f32-to-bf16",
743
+ "nbytes": 31457280,
744
+ "byteOffset": 0
745
+ }
746
+ ],
747
+ "md5sum": "56b6d3ede5ea23c9994fb02c48a82cd7"
748
+ },
749
+ {
750
+ "dataPath": "params_shard_39.bin",
751
+ "format": "raw-shard",
752
+ "nbytes": 18874368,
753
+ "records": [
754
+ {
755
+ "name": "model.layers.17.self_attn.o_proj.weight",
756
+ "shape": [
757
+ 3072,
758
+ 3072
759
+ ],
760
+ "dtype": "float32",
761
+ "format": "f32-to-bf16",
762
+ "nbytes": 18874368,
763
+ "byteOffset": 0
764
+ }
765
+ ],
766
+ "md5sum": "f6c60380deb7a8afbed11b307b145e18"
767
+ },
768
+ {
769
+ "dataPath": "params_shard_40.bin",
770
+ "format": "raw-shard",
771
+ "nbytes": 50331648,
772
+ "records": [
773
+ {
774
+ "name": "model.layers.18.mlp.down_proj.weight",
775
+ "shape": [
776
+ 3072,
777
+ 8192
778
+ ],
779
+ "dtype": "float32",
780
+ "format": "f32-to-bf16",
781
+ "nbytes": 50331648,
782
+ "byteOffset": 0
783
+ }
784
+ ],
785
+ "md5sum": "088c2b1caad88b346fbb5a3cb0d45110"
786
+ },
787
+ {
788
+ "dataPath": "params_shard_41.bin",
789
+ "format": "raw-shard",
790
+ "nbytes": 100663296,
791
+ "records": [
792
+ {
793
+ "name": "model.layers.18.mlp.gate_up_proj.weight",
794
+ "shape": [
795
+ 16384,
796
+ 3072
797
+ ],
798
+ "dtype": "float32",
799
+ "format": "f32-to-bf16",
800
+ "nbytes": 100663296,
801
+ "byteOffset": 0
802
+ }
803
+ ],
804
+ "md5sum": "44d02bd8483c1a0941af26e32438bb55"
805
+ },
806
+ {
807
+ "dataPath": "params_shard_42.bin",
808
+ "format": "raw-shard",
809
+ "nbytes": 31457280,
810
+ "records": [
811
+ {
812
+ "name": "model.layers.18.self_attn.qkv_proj.weight",
813
+ "shape": [
814
+ 5120,
815
+ 3072
816
+ ],
817
+ "dtype": "float32",
818
+ "format": "f32-to-bf16",
819
+ "nbytes": 31457280,
820
+ "byteOffset": 0
821
+ }
822
+ ],
823
+ "md5sum": "824146de15eb8206b9e7531a3a068b6f"
824
+ },
825
+ {
826
+ "dataPath": "params_shard_43.bin",
827
+ "format": "raw-shard",
828
+ "nbytes": 18874368,
829
+ "records": [
830
+ {
831
+ "name": "model.layers.18.self_attn.o_proj.weight",
832
+ "shape": [
833
+ 3072,
834
+ 3072
835
+ ],
836
+ "dtype": "float32",
837
+ "format": "f32-to-bf16",
838
+ "nbytes": 18874368,
839
+ "byteOffset": 0
840
+ }
841
+ ],
842
+ "md5sum": "ad1677d7e01e4ef22f342607f16f02df"
843
+ },
844
+ {
845
+ "dataPath": "params_shard_44.bin",
846
+ "format": "raw-shard",
847
+ "nbytes": 50331648,
848
+ "records": [
849
+ {
850
+ "name": "model.layers.19.mlp.down_proj.weight",
851
+ "shape": [
852
+ 3072,
853
+ 8192
854
+ ],
855
+ "dtype": "float32",
856
+ "format": "f32-to-bf16",
857
+ "nbytes": 50331648,
858
+ "byteOffset": 0
859
+ }
860
+ ],
861
+ "md5sum": "3143003000df5244f9fcfda00b62e2a0"
862
+ },
863
+ {
864
+ "dataPath": "params_shard_45.bin",
865
+ "format": "raw-shard",
866
+ "nbytes": 100663296,
867
+ "records": [
868
+ {
869
+ "name": "model.layers.19.mlp.gate_up_proj.weight",
870
+ "shape": [
871
+ 16384,
872
+ 3072
873
+ ],
874
+ "dtype": "float32",
875
+ "format": "f32-to-bf16",
876
+ "nbytes": 100663296,
877
+ "byteOffset": 0
878
+ }
879
+ ],
880
+ "md5sum": "96d5a23e28e0cd687ee129bf25bdd674"
881
+ },
882
+ {
883
+ "dataPath": "params_shard_46.bin",
884
+ "format": "raw-shard",
885
+ "nbytes": 31457280,
886
+ "records": [
887
+ {
888
+ "name": "model.layers.19.self_attn.qkv_proj.weight",
889
+ "shape": [
890
+ 5120,
891
+ 3072
892
+ ],
893
+ "dtype": "float32",
894
+ "format": "f32-to-bf16",
895
+ "nbytes": 31457280,
896
+ "byteOffset": 0
897
+ }
898
+ ],
899
+ "md5sum": "e7445744de4fce67eecd628cb5955f42"
900
+ },
901
+ {
902
+ "dataPath": "params_shard_47.bin",
903
+ "format": "raw-shard",
904
+ "nbytes": 18874368,
905
+ "records": [
906
+ {
907
+ "name": "model.layers.19.self_attn.o_proj.weight",
908
+ "shape": [
909
+ 3072,
910
+ 3072
911
+ ],
912
+ "dtype": "float32",
913
+ "format": "f32-to-bf16",
914
+ "nbytes": 18874368,
915
+ "byteOffset": 0
916
+ }
917
+ ],
918
+ "md5sum": "4e8a3a848bc7c572129bde24545e91ab"
919
+ },
920
+ {
921
+ "dataPath": "params_shard_48.bin",
922
+ "format": "raw-shard",
923
+ "nbytes": 50331648,
924
+ "records": [
925
+ {
926
+ "name": "model.layers.2.mlp.down_proj.weight",
927
+ "shape": [
928
+ 3072,
929
+ 8192
930
+ ],
931
+ "dtype": "float32",
932
+ "format": "f32-to-bf16",
933
+ "nbytes": 50331648,
934
+ "byteOffset": 0
935
+ }
936
+ ],
937
+ "md5sum": "c0d47846f1f268c6b9d034420139e267"
938
+ },
939
+ {
940
+ "dataPath": "params_shard_49.bin",
941
+ "format": "raw-shard",
942
+ "nbytes": 100663296,
943
+ "records": [
944
+ {
945
+ "name": "model.layers.2.mlp.gate_up_proj.weight",
946
+ "shape": [
947
+ 16384,
948
+ 3072
949
+ ],
950
+ "dtype": "float32",
951
+ "format": "f32-to-bf16",
952
+ "nbytes": 100663296,
953
+ "byteOffset": 0
954
+ }
955
+ ],
956
+ "md5sum": "ff5c419b4b570dfa26fe530cd0602a75"
957
+ },
958
+ {
959
+ "dataPath": "params_shard_50.bin",
960
+ "format": "raw-shard",
961
+ "nbytes": 31457280,
962
+ "records": [
963
+ {
964
+ "name": "model.layers.2.self_attn.qkv_proj.weight",
965
+ "shape": [
966
+ 5120,
967
+ 3072
968
+ ],
969
+ "dtype": "float32",
970
+ "format": "f32-to-bf16",
971
+ "nbytes": 31457280,
972
+ "byteOffset": 0
973
+ }
974
+ ],
975
+ "md5sum": "c11179f8247d6d4c364971a6450740a4"
976
+ },
977
+ {
978
+ "dataPath": "params_shard_51.bin",
979
+ "format": "raw-shard",
980
+ "nbytes": 18874368,
981
+ "records": [
982
+ {
983
+ "name": "model.layers.2.self_attn.o_proj.weight",
984
+ "shape": [
985
+ 3072,
986
+ 3072
987
+ ],
988
+ "dtype": "float32",
989
+ "format": "f32-to-bf16",
990
+ "nbytes": 18874368,
991
+ "byteOffset": 0
992
+ }
993
+ ],
994
+ "md5sum": "7588e5e7fa4304beb826fcaa6f3b3e65"
995
+ },
996
+ {
997
+ "dataPath": "params_shard_52.bin",
998
+ "format": "raw-shard",
999
+ "nbytes": 100663296,
1000
+ "records": [
1001
+ {
1002
+ "name": "model.layers.20.mlp.gate_up_proj.weight",
1003
+ "shape": [
1004
+ 16384,
1005
+ 3072
1006
+ ],
1007
+ "dtype": "float32",
1008
+ "format": "f32-to-bf16",
1009
+ "nbytes": 100663296,
1010
+ "byteOffset": 0
1011
+ }
1012
+ ],
1013
+ "md5sum": "49d7128811effec28ba14cd181946cb9"
1014
+ },
1015
+ {
1016
+ "dataPath": "params_shard_53.bin",
1017
+ "format": "raw-shard",
1018
+ "nbytes": 31457280,
1019
+ "records": [
1020
+ {
1021
+ "name": "model.layers.20.self_attn.qkv_proj.weight",
1022
+ "shape": [
1023
+ 5120,
1024
+ 3072
1025
+ ],
1026
+ "dtype": "float32",
1027
+ "format": "f32-to-bf16",
1028
+ "nbytes": 31457280,
1029
+ "byteOffset": 0
1030
+ }
1031
+ ],
1032
+ "md5sum": "b2559fc7223d8081ea4ff1f38efe170d"
1033
+ },
1034
+ {
1035
+ "dataPath": "params_shard_54.bin",
1036
+ "format": "raw-shard",
1037
+ "nbytes": 18874368,
1038
+ "records": [
1039
+ {
1040
+ "name": "model.layers.20.self_attn.o_proj.weight",
1041
+ "shape": [
1042
+ 3072,
1043
+ 3072
1044
+ ],
1045
+ "dtype": "float32",
1046
+ "format": "f32-to-bf16",
1047
+ "nbytes": 18874368,
1048
+ "byteOffset": 0
1049
+ }
1050
+ ],
1051
+ "md5sum": "427dbed3aa91d806dfcb213501ac507e"
1052
+ },
1053
+ {
1054
+ "dataPath": "params_shard_55.bin",
1055
+ "format": "raw-shard",
1056
+ "nbytes": 50331648,
1057
+ "records": [
1058
+ {
1059
+ "name": "model.layers.3.mlp.down_proj.weight",
1060
+ "shape": [
1061
+ 3072,
1062
+ 8192
1063
+ ],
1064
+ "dtype": "float32",
1065
+ "format": "f32-to-bf16",
1066
+ "nbytes": 50331648,
1067
+ "byteOffset": 0
1068
+ }
1069
+ ],
1070
+ "md5sum": "c86c1149c905eeca37abb6ace6b65081"
1071
+ },
1072
+ {
1073
+ "dataPath": "params_shard_56.bin",
1074
+ "format": "raw-shard",
1075
+ "nbytes": 100663296,
1076
+ "records": [
1077
+ {
1078
+ "name": "model.layers.3.mlp.gate_up_proj.weight",
1079
+ "shape": [
1080
+ 16384,
1081
+ 3072
1082
+ ],
1083
+ "dtype": "float32",
1084
+ "format": "f32-to-bf16",
1085
+ "nbytes": 100663296,
1086
+ "byteOffset": 0
1087
+ }
1088
+ ],
1089
+ "md5sum": "d0270175e869b30212326d8898f6bc6e"
1090
+ },
1091
+ {
1092
+ "dataPath": "params_shard_57.bin",
1093
+ "format": "raw-shard",
1094
+ "nbytes": 31457280,
1095
+ "records": [
1096
+ {
1097
+ "name": "model.layers.3.self_attn.qkv_proj.weight",
1098
+ "shape": [
1099
+ 5120,
1100
+ 3072
1101
+ ],
1102
+ "dtype": "float32",
1103
+ "format": "f32-to-bf16",
1104
+ "nbytes": 31457280,
1105
+ "byteOffset": 0
1106
+ }
1107
+ ],
1108
+ "md5sum": "97edc380dd2ad36b07b1dc01832a4407"
1109
+ },
1110
+ {
1111
+ "dataPath": "params_shard_58.bin",
1112
+ "format": "raw-shard",
1113
+ "nbytes": 18874368,
1114
+ "records": [
1115
+ {
1116
+ "name": "model.layers.3.self_attn.o_proj.weight",
1117
+ "shape": [
1118
+ 3072,
1119
+ 3072
1120
+ ],
1121
+ "dtype": "float32",
1122
+ "format": "f32-to-bf16",
1123
+ "nbytes": 18874368,
1124
+ "byteOffset": 0
1125
+ }
1126
+ ],
1127
+ "md5sum": "517ca02bf6261842f24247640bc0903f"
1128
+ },
1129
+ {
1130
+ "dataPath": "params_shard_59.bin",
1131
+ "format": "raw-shard",
1132
+ "nbytes": 50331648,
1133
+ "records": [
1134
+ {
1135
+ "name": "model.layers.4.mlp.down_proj.weight",
1136
+ "shape": [
1137
+ 3072,
1138
+ 8192
1139
+ ],
1140
+ "dtype": "float32",
1141
+ "format": "f32-to-bf16",
1142
+ "nbytes": 50331648,
1143
+ "byteOffset": 0
1144
+ }
1145
+ ],
1146
+ "md5sum": "7ae5988d6a252fc160f8afec618d78f8"
1147
+ },
1148
+ {
1149
+ "dataPath": "params_shard_60.bin",
1150
+ "format": "raw-shard",
1151
+ "nbytes": 100663296,
1152
+ "records": [
1153
+ {
1154
+ "name": "model.layers.4.mlp.gate_up_proj.weight",
1155
+ "shape": [
1156
+ 16384,
1157
+ 3072
1158
+ ],
1159
+ "dtype": "float32",
1160
+ "format": "f32-to-bf16",
1161
+ "nbytes": 100663296,
1162
+ "byteOffset": 0
1163
+ }
1164
+ ],
1165
+ "md5sum": "31d045e4bc48359a0a9355a6b1227faa"
1166
+ },
1167
+ {
1168
+ "dataPath": "params_shard_61.bin",
1169
+ "format": "raw-shard",
1170
+ "nbytes": 31457280,
1171
+ "records": [
1172
+ {
1173
+ "name": "model.layers.4.self_attn.qkv_proj.weight",
1174
+ "shape": [
1175
+ 5120,
1176
+ 3072
1177
+ ],
1178
+ "dtype": "float32",
1179
+ "format": "f32-to-bf16",
1180
+ "nbytes": 31457280,
1181
+ "byteOffset": 0
1182
+ }
1183
+ ],
1184
+ "md5sum": "2499d76e5b76de9a57ee472ec90d7385"
1185
+ },
1186
+ {
1187
+ "dataPath": "params_shard_62.bin",
1188
+ "format": "raw-shard",
1189
+ "nbytes": 18874368,
1190
+ "records": [
1191
+ {
1192
+ "name": "model.layers.4.self_attn.o_proj.weight",
1193
+ "shape": [
1194
+ 3072,
1195
+ 3072
1196
+ ],
1197
+ "dtype": "float32",
1198
+ "format": "f32-to-bf16",
1199
+ "nbytes": 18874368,
1200
+ "byteOffset": 0
1201
+ }
1202
+ ],
1203
+ "md5sum": "75c244d8197b1919723016fa125500b5"
1204
+ },
1205
+ {
1206
+ "dataPath": "params_shard_63.bin",
1207
+ "format": "raw-shard",
1208
+ "nbytes": 50331648,
1209
+ "records": [
1210
+ {
1211
+ "name": "model.layers.5.mlp.down_proj.weight",
1212
+ "shape": [
1213
+ 3072,
1214
+ 8192
1215
+ ],
1216
+ "dtype": "float32",
1217
+ "format": "f32-to-bf16",
1218
+ "nbytes": 50331648,
1219
+ "byteOffset": 0
1220
+ }
1221
+ ],
1222
+ "md5sum": "33c9ea14ac77d2ece67498ce22af6419"
1223
+ },
1224
+ {
1225
+ "dataPath": "params_shard_64.bin",
1226
+ "format": "raw-shard",
1227
+ "nbytes": 100663296,
1228
+ "records": [
1229
+ {
1230
+ "name": "model.layers.5.mlp.gate_up_proj.weight",
1231
+ "shape": [
1232
+ 16384,
1233
+ 3072
1234
+ ],
1235
+ "dtype": "float32",
1236
+ "format": "f32-to-bf16",
1237
+ "nbytes": 100663296,
1238
+ "byteOffset": 0
1239
+ }
1240
+ ],
1241
+ "md5sum": "5f901b9695815d535efed5b0a4a1e931"
1242
+ },
1243
+ {
1244
+ "dataPath": "params_shard_65.bin",
1245
+ "format": "raw-shard",
1246
+ "nbytes": 31457280,
1247
+ "records": [
1248
+ {
1249
+ "name": "model.layers.5.self_attn.qkv_proj.weight",
1250
+ "shape": [
1251
+ 5120,
1252
+ 3072
1253
+ ],
1254
+ "dtype": "float32",
1255
+ "format": "f32-to-bf16",
1256
+ "nbytes": 31457280,
1257
+ "byteOffset": 0
1258
+ }
1259
+ ],
1260
+ "md5sum": "d7806491318f8baaf9769a0bded6dc0b"
1261
+ },
1262
+ {
1263
+ "dataPath": "params_shard_66.bin",
1264
+ "format": "raw-shard",
1265
+ "nbytes": 18874368,
1266
+ "records": [
1267
+ {
1268
+ "name": "model.layers.5.self_attn.o_proj.weight",
1269
+ "shape": [
1270
+ 3072,
1271
+ 3072
1272
+ ],
1273
+ "dtype": "float32",
1274
+ "format": "f32-to-bf16",
1275
+ "nbytes": 18874368,
1276
+ "byteOffset": 0
1277
+ }
1278
+ ],
1279
+ "md5sum": "7d60368414ca9cab75b2ae03b15fb208"
1280
+ },
1281
+ {
1282
+ "dataPath": "params_shard_67.bin",
1283
+ "format": "raw-shard",
1284
+ "nbytes": 50331648,
1285
+ "records": [
1286
+ {
1287
+ "name": "model.layers.6.mlp.down_proj.weight",
1288
+ "shape": [
1289
+ 3072,
1290
+ 8192
1291
+ ],
1292
+ "dtype": "float32",
1293
+ "format": "f32-to-bf16",
1294
+ "nbytes": 50331648,
1295
+ "byteOffset": 0
1296
+ }
1297
+ ],
1298
+ "md5sum": "3052bcd3ac2b8d80ef17287f901bb3b9"
1299
+ },
1300
+ {
1301
+ "dataPath": "params_shard_68.bin",
1302
+ "format": "raw-shard",
1303
+ "nbytes": 100663296,
1304
+ "records": [
1305
+ {
1306
+ "name": "model.layers.6.mlp.gate_up_proj.weight",
1307
+ "shape": [
1308
+ 16384,
1309
+ 3072
1310
+ ],
1311
+ "dtype": "float32",
1312
+ "format": "f32-to-bf16",
1313
+ "nbytes": 100663296,
1314
+ "byteOffset": 0
1315
+ }
1316
+ ],
1317
+ "md5sum": "c511a17ee7127ea3514af602b5b5f75e"
1318
+ },
1319
+ {
1320
+ "dataPath": "params_shard_69.bin",
1321
+ "format": "raw-shard",
1322
+ "nbytes": 31457280,
1323
+ "records": [
1324
+ {
1325
+ "name": "model.layers.6.self_attn.qkv_proj.weight",
1326
+ "shape": [
1327
+ 5120,
1328
+ 3072
1329
+ ],
1330
+ "dtype": "float32",
1331
+ "format": "f32-to-bf16",
1332
+ "nbytes": 31457280,
1333
+ "byteOffset": 0
1334
+ }
1335
+ ],
1336
+ "md5sum": "e6e79670e5e719b3cec1e97f318218df"
1337
+ },
1338
+ {
1339
+ "dataPath": "params_shard_70.bin",
1340
+ "format": "raw-shard",
1341
+ "nbytes": 18874368,
1342
+ "records": [
1343
+ {
1344
+ "name": "model.layers.6.self_attn.o_proj.weight",
1345
+ "shape": [
1346
+ 3072,
1347
+ 3072
1348
+ ],
1349
+ "dtype": "float32",
1350
+ "format": "f32-to-bf16",
1351
+ "nbytes": 18874368,
1352
+ "byteOffset": 0
1353
+ }
1354
+ ],
1355
+ "md5sum": "90bb59b9defd080833ae995f838e39fc"
1356
+ },
1357
+ {
1358
+ "dataPath": "params_shard_71.bin",
1359
+ "format": "raw-shard",
1360
+ "nbytes": 50331648,
1361
+ "records": [
1362
+ {
1363
+ "name": "model.layers.7.mlp.down_proj.weight",
1364
+ "shape": [
1365
+ 3072,
1366
+ 8192
1367
+ ],
1368
+ "dtype": "float32",
1369
+ "format": "f32-to-bf16",
1370
+ "nbytes": 50331648,
1371
+ "byteOffset": 0
1372
+ }
1373
+ ],
1374
+ "md5sum": "6f7406fcbbaf37055796ee5282acd5e8"
1375
+ },
1376
+ {
1377
+ "dataPath": "params_shard_72.bin",
1378
+ "format": "raw-shard",
1379
+ "nbytes": 100663296,
1380
+ "records": [
1381
+ {
1382
+ "name": "model.layers.7.mlp.gate_up_proj.weight",
1383
+ "shape": [
1384
+ 16384,
1385
+ 3072
1386
+ ],
1387
+ "dtype": "float32",
1388
+ "format": "f32-to-bf16",
1389
+ "nbytes": 100663296,
1390
+ "byteOffset": 0
1391
+ }
1392
+ ],
1393
+ "md5sum": "d976299fe65efbcda8040d0bb0cdcff4"
1394
+ },
1395
+ {
1396
+ "dataPath": "params_shard_73.bin",
1397
+ "format": "raw-shard",
1398
+ "nbytes": 31457280,
1399
+ "records": [
1400
+ {
1401
+ "name": "model.layers.7.self_attn.qkv_proj.weight",
1402
+ "shape": [
1403
+ 5120,
1404
+ 3072
1405
+ ],
1406
+ "dtype": "float32",
1407
+ "format": "f32-to-bf16",
1408
+ "nbytes": 31457280,
1409
+ "byteOffset": 0
1410
+ }
1411
+ ],
1412
+ "md5sum": "1fcb92f405a175f578814f4f4c1ad88a"
1413
+ },
1414
+ {
1415
+ "dataPath": "params_shard_74.bin",
1416
+ "format": "raw-shard",
1417
+ "nbytes": 18874368,
1418
+ "records": [
1419
+ {
1420
+ "name": "model.layers.7.self_attn.o_proj.weight",
1421
+ "shape": [
1422
+ 3072,
1423
+ 3072
1424
+ ],
1425
+ "dtype": "float32",
1426
+ "format": "f32-to-bf16",
1427
+ "nbytes": 18874368,
1428
+ "byteOffset": 0
1429
+ }
1430
+ ],
1431
+ "md5sum": "71663fa01f184a407c2a5e19ec85e9e7"
1432
+ },
1433
+ {
1434
+ "dataPath": "params_shard_75.bin",
1435
+ "format": "raw-shard",
1436
+ "nbytes": 50331648,
1437
+ "records": [
1438
+ {
1439
+ "name": "model.layers.8.mlp.down_proj.weight",
1440
+ "shape": [
1441
+ 3072,
1442
+ 8192
1443
+ ],
1444
+ "dtype": "float32",
1445
+ "format": "f32-to-bf16",
1446
+ "nbytes": 50331648,
1447
+ "byteOffset": 0
1448
+ }
1449
+ ],
1450
+ "md5sum": "4e9ba681099e252c6cd123007b45e416"
1451
+ },
1452
+ {
1453
+ "dataPath": "params_shard_76.bin",
1454
+ "format": "raw-shard",
1455
+ "nbytes": 100663296,
1456
+ "records": [
1457
+ {
1458
+ "name": "model.layers.8.mlp.gate_up_proj.weight",
1459
+ "shape": [
1460
+ 16384,
1461
+ 3072
1462
+ ],
1463
+ "dtype": "float32",
1464
+ "format": "f32-to-bf16",
1465
+ "nbytes": 100663296,
1466
+ "byteOffset": 0
1467
+ }
1468
+ ],
1469
+ "md5sum": "e58dad56d1aa5aa3306cb3190ba7ff1f"
1470
+ },
1471
+ {
1472
+ "dataPath": "params_shard_77.bin",
1473
+ "format": "raw-shard",
1474
+ "nbytes": 31457280,
1475
+ "records": [
1476
+ {
1477
+ "name": "model.layers.8.self_attn.qkv_proj.weight",
1478
+ "shape": [
1479
+ 5120,
1480
+ 3072
1481
+ ],
1482
+ "dtype": "float32",
1483
+ "format": "f32-to-bf16",
1484
+ "nbytes": 31457280,
1485
+ "byteOffset": 0
1486
+ }
1487
+ ],
1488
+ "md5sum": "0001d2ba2d5eea1354aef2f5c727fe07"
1489
+ },
1490
+ {
1491
+ "dataPath": "params_shard_78.bin",
1492
+ "format": "raw-shard",
1493
+ "nbytes": 18874368,
1494
+ "records": [
1495
+ {
1496
+ "name": "model.layers.8.self_attn.o_proj.weight",
1497
+ "shape": [
1498
+ 3072,
1499
+ 3072
1500
+ ],
1501
+ "dtype": "float32",
1502
+ "format": "f32-to-bf16",
1503
+ "nbytes": 18874368,
1504
+ "byteOffset": 0
1505
+ }
1506
+ ],
1507
+ "md5sum": "76f9235ba682cf5b0388c94cdef2ec15"
1508
+ },
1509
+ {
1510
+ "dataPath": "params_shard_79.bin",
1511
+ "format": "raw-shard",
1512
+ "nbytes": 50331648,
1513
+ "records": [
1514
+ {
1515
+ "name": "model.layers.9.mlp.down_proj.weight",
1516
+ "shape": [
1517
+ 3072,
1518
+ 8192
1519
+ ],
1520
+ "dtype": "float32",
1521
+ "format": "f32-to-bf16",
1522
+ "nbytes": 50331648,
1523
+ "byteOffset": 0
1524
+ }
1525
+ ],
1526
+ "md5sum": "5b7ac8c47930806d061094840999d5d7"
1527
+ },
1528
+ {
1529
+ "dataPath": "params_shard_80.bin",
1530
+ "format": "raw-shard",
1531
+ "nbytes": 100663296,
1532
+ "records": [
1533
+ {
1534
+ "name": "model.layers.9.mlp.gate_up_proj.weight",
1535
+ "shape": [
1536
+ 16384,
1537
+ 3072
1538
+ ],
1539
+ "dtype": "float32",
1540
+ "format": "f32-to-bf16",
1541
+ "nbytes": 100663296,
1542
+ "byteOffset": 0
1543
+ }
1544
+ ],
1545
+ "md5sum": "02e74665c93354e4829fb04653abc27b"
1546
+ },
1547
+ {
1548
+ "dataPath": "params_shard_81.bin",
1549
+ "format": "raw-shard",
1550
+ "nbytes": 31457280,
1551
+ "records": [
1552
+ {
1553
+ "name": "model.layers.9.self_attn.qkv_proj.weight",
1554
+ "shape": [
1555
+ 5120,
1556
+ 3072
1557
+ ],
1558
+ "dtype": "float32",
1559
+ "format": "f32-to-bf16",
1560
+ "nbytes": 31457280,
1561
+ "byteOffset": 0
1562
+ }
1563
+ ],
1564
+ "md5sum": "ac1cff11d9b4cc142961891e4b0fb7de"
1565
+ },
1566
+ {
1567
+ "dataPath": "params_shard_82.bin",
1568
+ "format": "raw-shard",
1569
+ "nbytes": 18874368,
1570
+ "records": [
1571
+ {
1572
+ "name": "model.layers.9.self_attn.o_proj.weight",
1573
+ "shape": [
1574
+ 3072,
1575
+ 3072
1576
+ ],
1577
+ "dtype": "float32",
1578
+ "format": "f32-to-bf16",
1579
+ "nbytes": 18874368,
1580
+ "byteOffset": 0
1581
+ }
1582
+ ],
1583
+ "md5sum": "67bbc168157c53d1a1a1ff8b6052eaa1"
1584
+ },
1585
+ {
1586
+ "dataPath": "params_shard_83.bin",
1587
+ "format": "raw-shard",
1588
+ "nbytes": 50331648,
1589
+ "records": [
1590
+ {
1591
+ "name": "model.layers.20.mlp.down_proj.weight",
1592
+ "shape": [
1593
+ 3072,
1594
+ 8192
1595
+ ],
1596
+ "dtype": "float32",
1597
+ "format": "f32-to-bf16",
1598
+ "nbytes": 50331648,
1599
+ "byteOffset": 0
1600
+ }
1601
+ ],
1602
+ "md5sum": "cef4f79d43d9eace1ad5c2bc0bbab32c"
1603
+ },
1604
+ {
1605
+ "dataPath": "params_shard_84.bin",
1606
+ "format": "raw-shard",
1607
+ "nbytes": 50331648,
1608
+ "records": [
1609
+ {
1610
+ "name": "model.layers.21.mlp.down_proj.weight",
1611
+ "shape": [
1612
+ 3072,
1613
+ 8192
1614
+ ],
1615
+ "dtype": "float32",
1616
+ "format": "f32-to-bf16",
1617
+ "nbytes": 50331648,
1618
+ "byteOffset": 0
1619
+ }
1620
+ ],
1621
+ "md5sum": "caa538f23db663ddb35a0f7613bced7a"
1622
+ },
1623
+ {
1624
+ "dataPath": "params_shard_85.bin",
1625
+ "format": "raw-shard",
1626
+ "nbytes": 100663296,
1627
+ "records": [
1628
+ {
1629
+ "name": "model.layers.21.mlp.gate_up_proj.weight",
1630
+ "shape": [
1631
+ 16384,
1632
+ 3072
1633
+ ],
1634
+ "dtype": "float32",
1635
+ "format": "f32-to-bf16",
1636
+ "nbytes": 100663296,
1637
+ "byteOffset": 0
1638
+ }
1639
+ ],
1640
+ "md5sum": "ae2c3067f8024380c9729aeb23055920"
1641
+ },
1642
+ {
1643
+ "dataPath": "params_shard_86.bin",
1644
+ "format": "raw-shard",
1645
+ "nbytes": 31457280,
1646
+ "records": [
1647
+ {
1648
+ "name": "model.layers.21.self_attn.qkv_proj.weight",
1649
+ "shape": [
1650
+ 5120,
1651
+ 3072
1652
+ ],
1653
+ "dtype": "float32",
1654
+ "format": "f32-to-bf16",
1655
+ "nbytes": 31457280,
1656
+ "byteOffset": 0
1657
+ }
1658
+ ],
1659
+ "md5sum": "80fa2ae93a50724674d3ccf3097bc110"
1660
+ },
1661
+ {
1662
+ "dataPath": "params_shard_87.bin",
1663
+ "format": "raw-shard",
1664
+ "nbytes": 18874368,
1665
+ "records": [
1666
+ {
1667
+ "name": "model.layers.21.self_attn.o_proj.weight",
1668
+ "shape": [
1669
+ 3072,
1670
+ 3072
1671
+ ],
1672
+ "dtype": "float32",
1673
+ "format": "f32-to-bf16",
1674
+ "nbytes": 18874368,
1675
+ "byteOffset": 0
1676
+ }
1677
+ ],
1678
+ "md5sum": "8005bb547cfd8eed05260de00e01d305"
1679
+ },
1680
+ {
1681
+ "dataPath": "params_shard_88.bin",
1682
+ "format": "raw-shard",
1683
+ "nbytes": 50331648,
1684
+ "records": [
1685
+ {
1686
+ "name": "model.layers.22.mlp.down_proj.weight",
1687
+ "shape": [
1688
+ 3072,
1689
+ 8192
1690
+ ],
1691
+ "dtype": "float32",
1692
+ "format": "f32-to-bf16",
1693
+ "nbytes": 50331648,
1694
+ "byteOffset": 0
1695
+ }
1696
+ ],
1697
+ "md5sum": "55d1db437255049a98db5ca1d677dd25"
1698
+ },
1699
+ {
1700
+ "dataPath": "params_shard_89.bin",
1701
+ "format": "raw-shard",
1702
+ "nbytes": 100663296,
1703
+ "records": [
1704
+ {
1705
+ "name": "model.layers.22.mlp.gate_up_proj.weight",
1706
+ "shape": [
1707
+ 16384,
1708
+ 3072
1709
+ ],
1710
+ "dtype": "float32",
1711
+ "format": "f32-to-bf16",
1712
+ "nbytes": 100663296,
1713
+ "byteOffset": 0
1714
+ }
1715
+ ],
1716
+ "md5sum": "e7f1f30f38670d683112ba5706a2f092"
1717
+ },
1718
+ {
1719
+ "dataPath": "params_shard_90.bin",
1720
+ "format": "raw-shard",
1721
+ "nbytes": 31457280,
1722
+ "records": [
1723
+ {
1724
+ "name": "model.layers.22.self_attn.qkv_proj.weight",
1725
+ "shape": [
1726
+ 5120,
1727
+ 3072
1728
+ ],
1729
+ "dtype": "float32",
1730
+ "format": "f32-to-bf16",
1731
+ "nbytes": 31457280,
1732
+ "byteOffset": 0
1733
+ }
1734
+ ],
1735
+ "md5sum": "04a985de9da66349a9e0401546d9a065"
1736
+ },
1737
+ {
1738
+ "dataPath": "params_shard_91.bin",
1739
+ "format": "raw-shard",
1740
+ "nbytes": 18874368,
1741
+ "records": [
1742
+ {
1743
+ "name": "model.layers.22.self_attn.o_proj.weight",
1744
+ "shape": [
1745
+ 3072,
1746
+ 3072
1747
+ ],
1748
+ "dtype": "float32",
1749
+ "format": "f32-to-bf16",
1750
+ "nbytes": 18874368,
1751
+ "byteOffset": 0
1752
+ }
1753
+ ],
1754
+ "md5sum": "65ecf4f35052edf63f9f3d20c470c5dd"
1755
+ },
1756
+ {
1757
+ "dataPath": "params_shard_92.bin",
1758
+ "format": "raw-shard",
1759
+ "nbytes": 50331648,
1760
+ "records": [
1761
+ {
1762
+ "name": "model.layers.23.mlp.down_proj.weight",
1763
+ "shape": [
1764
+ 3072,
1765
+ 8192
1766
+ ],
1767
+ "dtype": "float32",
1768
+ "format": "f32-to-bf16",
1769
+ "nbytes": 50331648,
1770
+ "byteOffset": 0
1771
+ }
1772
+ ],
1773
+ "md5sum": "f0d1c8ba2a582fdcc0e5270f7c061471"
1774
+ },
1775
+ {
1776
+ "dataPath": "params_shard_93.bin",
1777
+ "format": "raw-shard",
1778
+ "nbytes": 100663296,
1779
+ "records": [
1780
+ {
1781
+ "name": "model.layers.23.mlp.gate_up_proj.weight",
1782
+ "shape": [
1783
+ 16384,
1784
+ 3072
1785
+ ],
1786
+ "dtype": "float32",
1787
+ "format": "f32-to-bf16",
1788
+ "nbytes": 100663296,
1789
+ "byteOffset": 0
1790
+ }
1791
+ ],
1792
+ "md5sum": "a0a933b90f40ea9b50fdcde46463cd55"
1793
+ },
1794
+ {
1795
+ "dataPath": "params_shard_94.bin",
1796
+ "format": "raw-shard",
1797
+ "nbytes": 31457280,
1798
+ "records": [
1799
+ {
1800
+ "name": "model.layers.23.self_attn.qkv_proj.weight",
1801
+ "shape": [
1802
+ 5120,
1803
+ 3072
1804
+ ],
1805
+ "dtype": "float32",
1806
+ "format": "f32-to-bf16",
1807
+ "nbytes": 31457280,
1808
+ "byteOffset": 0
1809
+ }
1810
+ ],
1811
+ "md5sum": "da99b4fb689c104ee62ad2cad0cfa534"
1812
+ },
1813
+ {
1814
+ "dataPath": "params_shard_95.bin",
1815
+ "format": "raw-shard",
1816
+ "nbytes": 18874368,
1817
+ "records": [
1818
+ {
1819
+ "name": "model.layers.23.self_attn.o_proj.weight",
1820
+ "shape": [
1821
+ 3072,
1822
+ 3072
1823
+ ],
1824
+ "dtype": "float32",
1825
+ "format": "f32-to-bf16",
1826
+ "nbytes": 18874368,
1827
+ "byteOffset": 0
1828
+ }
1829
+ ],
1830
+ "md5sum": "0208161744636162aaddd958458c42f4"
1831
+ },
1832
+ {
1833
+ "dataPath": "params_shard_96.bin",
1834
+ "format": "raw-shard",
1835
+ "nbytes": 50331648,
1836
+ "records": [
1837
+ {
1838
+ "name": "model.layers.24.mlp.down_proj.weight",
1839
+ "shape": [
1840
+ 3072,
1841
+ 8192
1842
+ ],
1843
+ "dtype": "float32",
1844
+ "format": "f32-to-bf16",
1845
+ "nbytes": 50331648,
1846
+ "byteOffset": 0
1847
+ }
1848
+ ],
1849
+ "md5sum": "1c87befe152239ad5a565da807630695"
1850
+ },
1851
+ {
1852
+ "dataPath": "params_shard_97.bin",
1853
+ "format": "raw-shard",
1854
+ "nbytes": 100663296,
1855
+ "records": [
1856
+ {
1857
+ "name": "model.layers.24.mlp.gate_up_proj.weight",
1858
+ "shape": [
1859
+ 16384,
1860
+ 3072
1861
+ ],
1862
+ "dtype": "float32",
1863
+ "format": "f32-to-bf16",
1864
+ "nbytes": 100663296,
1865
+ "byteOffset": 0
1866
+ }
1867
+ ],
1868
+ "md5sum": "f9a41904bae1070952eabe69083f82d1"
1869
+ },
1870
+ {
1871
+ "dataPath": "params_shard_98.bin",
1872
+ "format": "raw-shard",
1873
+ "nbytes": 31457280,
1874
+ "records": [
1875
+ {
1876
+ "name": "model.layers.24.self_attn.qkv_proj.weight",
1877
+ "shape": [
1878
+ 5120,
1879
+ 3072
1880
+ ],
1881
+ "dtype": "float32",
1882
+ "format": "f32-to-bf16",
1883
+ "nbytes": 31457280,
1884
+ "byteOffset": 0
1885
+ }
1886
+ ],
1887
+ "md5sum": "3aed6dd9398fa605f8c3539853ba96fa"
1888
+ },
1889
+ {
1890
+ "dataPath": "params_shard_99.bin",
1891
+ "format": "raw-shard",
1892
+ "nbytes": 18874368,
1893
+ "records": [
1894
+ {
1895
+ "name": "model.layers.24.self_attn.o_proj.weight",
1896
+ "shape": [
1897
+ 3072,
1898
+ 3072
1899
+ ],
1900
+ "dtype": "float32",
1901
+ "format": "f32-to-bf16",
1902
+ "nbytes": 18874368,
1903
+ "byteOffset": 0
1904
+ }
1905
+ ],
1906
+ "md5sum": "6e3573ab844d10a115a672e5eb4101a1"
1907
+ },
1908
+ {
1909
+ "dataPath": "params_shard_100.bin",
1910
+ "format": "raw-shard",
1911
+ "nbytes": 50331648,
1912
+ "records": [
1913
+ {
1914
+ "name": "model.layers.25.mlp.down_proj.weight",
1915
+ "shape": [
1916
+ 3072,
1917
+ 8192
1918
+ ],
1919
+ "dtype": "float32",
1920
+ "format": "f32-to-bf16",
1921
+ "nbytes": 50331648,
1922
+ "byteOffset": 0
1923
+ }
1924
+ ],
1925
+ "md5sum": "1f0c9184fc17fa72a166b741fbd14304"
1926
+ },
1927
+ {
1928
+ "dataPath": "params_shard_101.bin",
1929
+ "format": "raw-shard",
1930
+ "nbytes": 100663296,
1931
+ "records": [
1932
+ {
1933
+ "name": "model.layers.25.mlp.gate_up_proj.weight",
1934
+ "shape": [
1935
+ 16384,
1936
+ 3072
1937
+ ],
1938
+ "dtype": "float32",
1939
+ "format": "f32-to-bf16",
1940
+ "nbytes": 100663296,
1941
+ "byteOffset": 0
1942
+ }
1943
+ ],
1944
+ "md5sum": "4e952489e44feff3f51f293243193963"
1945
+ },
1946
+ {
1947
+ "dataPath": "params_shard_102.bin",
1948
+ "format": "raw-shard",
1949
+ "nbytes": 31457280,
1950
+ "records": [
1951
+ {
1952
+ "name": "model.layers.25.self_attn.qkv_proj.weight",
1953
+ "shape": [
1954
+ 5120,
1955
+ 3072
1956
+ ],
1957
+ "dtype": "float32",
1958
+ "format": "f32-to-bf16",
1959
+ "nbytes": 31457280,
1960
+ "byteOffset": 0
1961
+ }
1962
+ ],
1963
+ "md5sum": "ac0f2806df1ac7868d2e83026271f057"
1964
+ },
1965
+ {
1966
+ "dataPath": "params_shard_103.bin",
1967
+ "format": "raw-shard",
1968
+ "nbytes": 18874368,
1969
+ "records": [
1970
+ {
1971
+ "name": "model.layers.25.self_attn.o_proj.weight",
1972
+ "shape": [
1973
+ 3072,
1974
+ 3072
1975
+ ],
1976
+ "dtype": "float32",
1977
+ "format": "f32-to-bf16",
1978
+ "nbytes": 18874368,
1979
+ "byteOffset": 0
1980
+ }
1981
+ ],
1982
+ "md5sum": "81c614dbc2fac010a3809d3c51055227"
1983
+ },
1984
+ {
1985
+ "dataPath": "params_shard_104.bin",
1986
+ "format": "raw-shard",
1987
+ "nbytes": 50331648,
1988
+ "records": [
1989
+ {
1990
+ "name": "model.layers.26.mlp.down_proj.weight",
1991
+ "shape": [
1992
+ 3072,
1993
+ 8192
1994
+ ],
1995
+ "dtype": "float32",
1996
+ "format": "f32-to-bf16",
1997
+ "nbytes": 50331648,
1998
+ "byteOffset": 0
1999
+ }
2000
+ ],
2001
+ "md5sum": "028de1825559128df8e5fd233404f13c"
2002
+ },
2003
+ {
2004
+ "dataPath": "params_shard_105.bin",
2005
+ "format": "raw-shard",
2006
+ "nbytes": 100663296,
2007
+ "records": [
2008
+ {
2009
+ "name": "model.layers.26.mlp.gate_up_proj.weight",
2010
+ "shape": [
2011
+ 16384,
2012
+ 3072
2013
+ ],
2014
+ "dtype": "float32",
2015
+ "format": "f32-to-bf16",
2016
+ "nbytes": 100663296,
2017
+ "byteOffset": 0
2018
+ }
2019
+ ],
2020
+ "md5sum": "91c6cc596728b4d45f07bf90e4ce40dc"
2021
+ },
2022
+ {
2023
+ "dataPath": "params_shard_106.bin",
2024
+ "format": "raw-shard",
2025
+ "nbytes": 31457280,
2026
+ "records": [
2027
+ {
2028
+ "name": "model.layers.26.self_attn.qkv_proj.weight",
2029
+ "shape": [
2030
+ 5120,
2031
+ 3072
2032
+ ],
2033
+ "dtype": "float32",
2034
+ "format": "f32-to-bf16",
2035
+ "nbytes": 31457280,
2036
+ "byteOffset": 0
2037
+ }
2038
+ ],
2039
+ "md5sum": "ef94f2bf8dd5bcbca41d8c1cdf0a9208"
2040
+ },
2041
+ {
2042
+ "dataPath": "params_shard_107.bin",
2043
+ "format": "raw-shard",
2044
+ "nbytes": 18874368,
2045
+ "records": [
2046
+ {
2047
+ "name": "model.layers.26.self_attn.o_proj.weight",
2048
+ "shape": [
2049
+ 3072,
2050
+ 3072
2051
+ ],
2052
+ "dtype": "float32",
2053
+ "format": "f32-to-bf16",
2054
+ "nbytes": 18874368,
2055
+ "byteOffset": 0
2056
+ }
2057
+ ],
2058
+ "md5sum": "403dbf175a8c9eec0e8f94d700500be8"
2059
+ },
2060
+ {
2061
+ "dataPath": "params_shard_108.bin",
2062
+ "format": "raw-shard",
2063
+ "nbytes": 50331648,
2064
+ "records": [
2065
+ {
2066
+ "name": "model.layers.27.mlp.down_proj.weight",
2067
+ "shape": [
2068
+ 3072,
2069
+ 8192
2070
+ ],
2071
+ "dtype": "float32",
2072
+ "format": "f32-to-bf16",
2073
+ "nbytes": 50331648,
2074
+ "byteOffset": 0
2075
+ }
2076
+ ],
2077
+ "md5sum": "d7db1850f8f194d48ce5fca0cb8aaf7d"
2078
+ },
2079
+ {
2080
+ "dataPath": "params_shard_109.bin",
2081
+ "format": "raw-shard",
2082
+ "nbytes": 100663296,
2083
+ "records": [
2084
+ {
2085
+ "name": "model.layers.27.mlp.gate_up_proj.weight",
2086
+ "shape": [
2087
+ 16384,
2088
+ 3072
2089
+ ],
2090
+ "dtype": "float32",
2091
+ "format": "f32-to-bf16",
2092
+ "nbytes": 100663296,
2093
+ "byteOffset": 0
2094
+ }
2095
+ ],
2096
+ "md5sum": "d970e24d360046aada8fa036e74a0104"
2097
+ },
2098
+ {
2099
+ "dataPath": "params_shard_110.bin",
2100
+ "format": "raw-shard",
2101
+ "nbytes": 31457280,
2102
+ "records": [
2103
+ {
2104
+ "name": "model.layers.27.self_attn.qkv_proj.weight",
2105
+ "shape": [
2106
+ 5120,
2107
+ 3072
2108
+ ],
2109
+ "dtype": "float32",
2110
+ "format": "f32-to-bf16",
2111
+ "nbytes": 31457280,
2112
+ "byteOffset": 0
2113
+ }
2114
+ ],
2115
+ "md5sum": "71d768ddcd341542e6b40db9ae197d46"
2116
+ },
2117
+ {
2118
+ "dataPath": "params_shard_111.bin",
2119
+ "format": "raw-shard",
2120
+ "nbytes": 18874368,
2121
+ "records": [
2122
+ {
2123
+ "name": "model.layers.27.self_attn.o_proj.weight",
2124
+ "shape": [
2125
+ 3072,
2126
+ 3072
2127
+ ],
2128
+ "dtype": "float32",
2129
+ "format": "f32-to-bf16",
2130
+ "nbytes": 18874368,
2131
+ "byteOffset": 0
2132
+ }
2133
+ ],
2134
+ "md5sum": "bf42c4b4af143df9f6def1800dd32225"
2135
+ },
2136
+ {
2137
+ "dataPath": "params_shard_112.bin",
2138
+ "format": "raw-shard",
2139
+ "nbytes": 31807488,
2140
+ "records": [
2141
+ {
2142
+ "name": "model.layers.0.input_layernorm.weight",
2143
+ "shape": [
2144
+ 3072
2145
+ ],
2146
+ "dtype": "float32",
2147
+ "format": "f32-to-bf16",
2148
+ "nbytes": 6144,
2149
+ "byteOffset": 0
2150
+ },
2151
+ {
2152
+ "name": "model.layers.0.post_attention_layernorm.weight",
2153
+ "shape": [
2154
+ 3072
2155
+ ],
2156
+ "dtype": "float32",
2157
+ "format": "f32-to-bf16",
2158
+ "nbytes": 6144,
2159
+ "byteOffset": 6144
2160
+ },
2161
+ {
2162
+ "name": "model.layers.0.self_attn.qkv_proj.weight",
2163
+ "shape": [
2164
+ 5120,
2165
+ 3072
2166
+ ],
2167
+ "dtype": "float32",
2168
+ "format": "f32-to-bf16",
2169
+ "nbytes": 31457280,
2170
+ "byteOffset": 12288
2171
+ },
2172
+ {
2173
+ "name": "model.layers.1.input_layernorm.weight",
2174
+ "shape": [
2175
+ 3072
2176
+ ],
2177
+ "dtype": "float32",
2178
+ "format": "f32-to-bf16",
2179
+ "nbytes": 6144,
2180
+ "byteOffset": 31469568
2181
+ },
2182
+ {
2183
+ "name": "model.layers.1.post_attention_layernorm.weight",
2184
+ "shape": [
2185
+ 3072
2186
+ ],
2187
+ "dtype": "float32",
2188
+ "format": "f32-to-bf16",
2189
+ "nbytes": 6144,
2190
+ "byteOffset": 31475712
2191
+ },
2192
+ {
2193
+ "name": "model.layers.10.input_layernorm.weight",
2194
+ "shape": [
2195
+ 3072
2196
+ ],
2197
+ "dtype": "float32",
2198
+ "format": "f32-to-bf16",
2199
+ "nbytes": 6144,
2200
+ "byteOffset": 31481856
2201
+ },
2202
+ {
2203
+ "name": "model.layers.10.post_attention_layernorm.weight",
2204
+ "shape": [
2205
+ 3072
2206
+ ],
2207
+ "dtype": "float32",
2208
+ "format": "f32-to-bf16",
2209
+ "nbytes": 6144,
2210
+ "byteOffset": 31488000
2211
+ },
2212
+ {
2213
+ "name": "model.layers.11.input_layernorm.weight",
2214
+ "shape": [
2215
+ 3072
2216
+ ],
2217
+ "dtype": "float32",
2218
+ "format": "f32-to-bf16",
2219
+ "nbytes": 6144,
2220
+ "byteOffset": 31494144
2221
+ },
2222
+ {
2223
+ "name": "model.layers.11.post_attention_layernorm.weight",
2224
+ "shape": [
2225
+ 3072
2226
+ ],
2227
+ "dtype": "float32",
2228
+ "format": "f32-to-bf16",
2229
+ "nbytes": 6144,
2230
+ "byteOffset": 31500288
2231
+ },
2232
+ {
2233
+ "name": "model.layers.12.input_layernorm.weight",
2234
+ "shape": [
2235
+ 3072
2236
+ ],
2237
+ "dtype": "float32",
2238
+ "format": "f32-to-bf16",
2239
+ "nbytes": 6144,
2240
+ "byteOffset": 31506432
2241
+ },
2242
+ {
2243
+ "name": "model.layers.12.post_attention_layernorm.weight",
2244
+ "shape": [
2245
+ 3072
2246
+ ],
2247
+ "dtype": "float32",
2248
+ "format": "f32-to-bf16",
2249
+ "nbytes": 6144,
2250
+ "byteOffset": 31512576
2251
+ },
2252
+ {
2253
+ "name": "model.layers.13.input_layernorm.weight",
2254
+ "shape": [
2255
+ 3072
2256
+ ],
2257
+ "dtype": "float32",
2258
+ "format": "f32-to-bf16",
2259
+ "nbytes": 6144,
2260
+ "byteOffset": 31518720
2261
+ },
2262
+ {
2263
+ "name": "model.layers.13.post_attention_layernorm.weight",
2264
+ "shape": [
2265
+ 3072
2266
+ ],
2267
+ "dtype": "float32",
2268
+ "format": "f32-to-bf16",
2269
+ "nbytes": 6144,
2270
+ "byteOffset": 31524864
2271
+ },
2272
+ {
2273
+ "name": "model.layers.14.input_layernorm.weight",
2274
+ "shape": [
2275
+ 3072
2276
+ ],
2277
+ "dtype": "float32",
2278
+ "format": "f32-to-bf16",
2279
+ "nbytes": 6144,
2280
+ "byteOffset": 31531008
2281
+ },
2282
+ {
2283
+ "name": "model.layers.14.post_attention_layernorm.weight",
2284
+ "shape": [
2285
+ 3072
2286
+ ],
2287
+ "dtype": "float32",
2288
+ "format": "f32-to-bf16",
2289
+ "nbytes": 6144,
2290
+ "byteOffset": 31537152
2291
+ },
2292
+ {
2293
+ "name": "model.layers.15.input_layernorm.weight",
2294
+ "shape": [
2295
+ 3072
2296
+ ],
2297
+ "dtype": "float32",
2298
+ "format": "f32-to-bf16",
2299
+ "nbytes": 6144,
2300
+ "byteOffset": 31543296
2301
+ },
2302
+ {
2303
+ "name": "model.layers.15.post_attention_layernorm.weight",
2304
+ "shape": [
2305
+ 3072
2306
+ ],
2307
+ "dtype": "float32",
2308
+ "format": "f32-to-bf16",
2309
+ "nbytes": 6144,
2310
+ "byteOffset": 31549440
2311
+ },
2312
+ {
2313
+ "name": "model.layers.16.input_layernorm.weight",
2314
+ "shape": [
2315
+ 3072
2316
+ ],
2317
+ "dtype": "float32",
2318
+ "format": "f32-to-bf16",
2319
+ "nbytes": 6144,
2320
+ "byteOffset": 31555584
2321
+ },
2322
+ {
2323
+ "name": "model.layers.16.post_attention_layernorm.weight",
2324
+ "shape": [
2325
+ 3072
2326
+ ],
2327
+ "dtype": "float32",
2328
+ "format": "f32-to-bf16",
2329
+ "nbytes": 6144,
2330
+ "byteOffset": 31561728
2331
+ },
2332
+ {
2333
+ "name": "model.layers.17.input_layernorm.weight",
2334
+ "shape": [
2335
+ 3072
2336
+ ],
2337
+ "dtype": "float32",
2338
+ "format": "f32-to-bf16",
2339
+ "nbytes": 6144,
2340
+ "byteOffset": 31567872
2341
+ },
2342
+ {
2343
+ "name": "model.layers.17.post_attention_layernorm.weight",
2344
+ "shape": [
2345
+ 3072
2346
+ ],
2347
+ "dtype": "float32",
2348
+ "format": "f32-to-bf16",
2349
+ "nbytes": 6144,
2350
+ "byteOffset": 31574016
2351
+ },
2352
+ {
2353
+ "name": "model.layers.18.input_layernorm.weight",
2354
+ "shape": [
2355
+ 3072
2356
+ ],
2357
+ "dtype": "float32",
2358
+ "format": "f32-to-bf16",
2359
+ "nbytes": 6144,
2360
+ "byteOffset": 31580160
2361
+ },
2362
+ {
2363
+ "name": "model.layers.18.post_attention_layernorm.weight",
2364
+ "shape": [
2365
+ 3072
2366
+ ],
2367
+ "dtype": "float32",
2368
+ "format": "f32-to-bf16",
2369
+ "nbytes": 6144,
2370
+ "byteOffset": 31586304
2371
+ },
2372
+ {
2373
+ "name": "model.layers.19.input_layernorm.weight",
2374
+ "shape": [
2375
+ 3072
2376
+ ],
2377
+ "dtype": "float32",
2378
+ "format": "f32-to-bf16",
2379
+ "nbytes": 6144,
2380
+ "byteOffset": 31592448
2381
+ },
2382
+ {
2383
+ "name": "model.layers.19.post_attention_layernorm.weight",
2384
+ "shape": [
2385
+ 3072
2386
+ ],
2387
+ "dtype": "float32",
2388
+ "format": "f32-to-bf16",
2389
+ "nbytes": 6144,
2390
+ "byteOffset": 31598592
2391
+ },
2392
+ {
2393
+ "name": "model.layers.2.input_layernorm.weight",
2394
+ "shape": [
2395
+ 3072
2396
+ ],
2397
+ "dtype": "float32",
2398
+ "format": "f32-to-bf16",
2399
+ "nbytes": 6144,
2400
+ "byteOffset": 31604736
2401
+ },
2402
+ {
2403
+ "name": "model.layers.2.post_attention_layernorm.weight",
2404
+ "shape": [
2405
+ 3072
2406
+ ],
2407
+ "dtype": "float32",
2408
+ "format": "f32-to-bf16",
2409
+ "nbytes": 6144,
2410
+ "byteOffset": 31610880
2411
+ },
2412
+ {
2413
+ "name": "model.layers.3.input_layernorm.weight",
2414
+ "shape": [
2415
+ 3072
2416
+ ],
2417
+ "dtype": "float32",
2418
+ "format": "f32-to-bf16",
2419
+ "nbytes": 6144,
2420
+ "byteOffset": 31617024
2421
+ },
2422
+ {
2423
+ "name": "model.layers.3.post_attention_layernorm.weight",
2424
+ "shape": [
2425
+ 3072
2426
+ ],
2427
+ "dtype": "float32",
2428
+ "format": "f32-to-bf16",
2429
+ "nbytes": 6144,
2430
+ "byteOffset": 31623168
2431
+ },
2432
+ {
2433
+ "name": "model.layers.4.input_layernorm.weight",
2434
+ "shape": [
2435
+ 3072
2436
+ ],
2437
+ "dtype": "float32",
2438
+ "format": "f32-to-bf16",
2439
+ "nbytes": 6144,
2440
+ "byteOffset": 31629312
2441
+ },
2442
+ {
2443
+ "name": "model.layers.4.post_attention_layernorm.weight",
2444
+ "shape": [
2445
+ 3072
2446
+ ],
2447
+ "dtype": "float32",
2448
+ "format": "f32-to-bf16",
2449
+ "nbytes": 6144,
2450
+ "byteOffset": 31635456
2451
+ },
2452
+ {
2453
+ "name": "model.layers.5.input_layernorm.weight",
2454
+ "shape": [
2455
+ 3072
2456
+ ],
2457
+ "dtype": "float32",
2458
+ "format": "f32-to-bf16",
2459
+ "nbytes": 6144,
2460
+ "byteOffset": 31641600
2461
+ },
2462
+ {
2463
+ "name": "model.layers.5.post_attention_layernorm.weight",
2464
+ "shape": [
2465
+ 3072
2466
+ ],
2467
+ "dtype": "float32",
2468
+ "format": "f32-to-bf16",
2469
+ "nbytes": 6144,
2470
+ "byteOffset": 31647744
2471
+ },
2472
+ {
2473
+ "name": "model.layers.6.input_layernorm.weight",
2474
+ "shape": [
2475
+ 3072
2476
+ ],
2477
+ "dtype": "float32",
2478
+ "format": "f32-to-bf16",
2479
+ "nbytes": 6144,
2480
+ "byteOffset": 31653888
2481
+ },
2482
+ {
2483
+ "name": "model.layers.6.post_attention_layernorm.weight",
2484
+ "shape": [
2485
+ 3072
2486
+ ],
2487
+ "dtype": "float32",
2488
+ "format": "f32-to-bf16",
2489
+ "nbytes": 6144,
2490
+ "byteOffset": 31660032
2491
+ },
2492
+ {
2493
+ "name": "model.layers.7.input_layernorm.weight",
2494
+ "shape": [
2495
+ 3072
2496
+ ],
2497
+ "dtype": "float32",
2498
+ "format": "f32-to-bf16",
2499
+ "nbytes": 6144,
2500
+ "byteOffset": 31666176
2501
+ },
2502
+ {
2503
+ "name": "model.layers.7.post_attention_layernorm.weight",
2504
+ "shape": [
2505
+ 3072
2506
+ ],
2507
+ "dtype": "float32",
2508
+ "format": "f32-to-bf16",
2509
+ "nbytes": 6144,
2510
+ "byteOffset": 31672320
2511
+ },
2512
+ {
2513
+ "name": "model.layers.8.input_layernorm.weight",
2514
+ "shape": [
2515
+ 3072
2516
+ ],
2517
+ "dtype": "float32",
2518
+ "format": "f32-to-bf16",
2519
+ "nbytes": 6144,
2520
+ "byteOffset": 31678464
2521
+ },
2522
+ {
2523
+ "name": "model.layers.8.post_attention_layernorm.weight",
2524
+ "shape": [
2525
+ 3072
2526
+ ],
2527
+ "dtype": "float32",
2528
+ "format": "f32-to-bf16",
2529
+ "nbytes": 6144,
2530
+ "byteOffset": 31684608
2531
+ },
2532
+ {
2533
+ "name": "model.layers.9.input_layernorm.weight",
2534
+ "shape": [
2535
+ 3072
2536
+ ],
2537
+ "dtype": "float32",
2538
+ "format": "f32-to-bf16",
2539
+ "nbytes": 6144,
2540
+ "byteOffset": 31690752
2541
+ },
2542
+ {
2543
+ "name": "model.layers.9.post_attention_layernorm.weight",
2544
+ "shape": [
2545
+ 3072
2546
+ ],
2547
+ "dtype": "float32",
2548
+ "format": "f32-to-bf16",
2549
+ "nbytes": 6144,
2550
+ "byteOffset": 31696896
2551
+ },
2552
+ {
2553
+ "name": "model.layers.20.input_layernorm.weight",
2554
+ "shape": [
2555
+ 3072
2556
+ ],
2557
+ "dtype": "float32",
2558
+ "format": "f32-to-bf16",
2559
+ "nbytes": 6144,
2560
+ "byteOffset": 31703040
2561
+ },
2562
+ {
2563
+ "name": "model.layers.20.post_attention_layernorm.weight",
2564
+ "shape": [
2565
+ 3072
2566
+ ],
2567
+ "dtype": "float32",
2568
+ "format": "f32-to-bf16",
2569
+ "nbytes": 6144,
2570
+ "byteOffset": 31709184
2571
+ },
2572
+ {
2573
+ "name": "model.layers.21.input_layernorm.weight",
2574
+ "shape": [
2575
+ 3072
2576
+ ],
2577
+ "dtype": "float32",
2578
+ "format": "f32-to-bf16",
2579
+ "nbytes": 6144,
2580
+ "byteOffset": 31715328
2581
+ },
2582
+ {
2583
+ "name": "model.layers.21.post_attention_layernorm.weight",
2584
+ "shape": [
2585
+ 3072
2586
+ ],
2587
+ "dtype": "float32",
2588
+ "format": "f32-to-bf16",
2589
+ "nbytes": 6144,
2590
+ "byteOffset": 31721472
2591
+ },
2592
+ {
2593
+ "name": "model.layers.22.input_layernorm.weight",
2594
+ "shape": [
2595
+ 3072
2596
+ ],
2597
+ "dtype": "float32",
2598
+ "format": "f32-to-bf16",
2599
+ "nbytes": 6144,
2600
+ "byteOffset": 31727616
2601
+ },
2602
+ {
2603
+ "name": "model.layers.22.post_attention_layernorm.weight",
2604
+ "shape": [
2605
+ 3072
2606
+ ],
2607
+ "dtype": "float32",
2608
+ "format": "f32-to-bf16",
2609
+ "nbytes": 6144,
2610
+ "byteOffset": 31733760
2611
+ },
2612
+ {
2613
+ "name": "model.layers.23.input_layernorm.weight",
2614
+ "shape": [
2615
+ 3072
2616
+ ],
2617
+ "dtype": "float32",
2618
+ "format": "f32-to-bf16",
2619
+ "nbytes": 6144,
2620
+ "byteOffset": 31739904
2621
+ },
2622
+ {
2623
+ "name": "model.layers.23.post_attention_layernorm.weight",
2624
+ "shape": [
2625
+ 3072
2626
+ ],
2627
+ "dtype": "float32",
2628
+ "format": "f32-to-bf16",
2629
+ "nbytes": 6144,
2630
+ "byteOffset": 31746048
2631
+ },
2632
+ {
2633
+ "name": "model.layers.24.input_layernorm.weight",
2634
+ "shape": [
2635
+ 3072
2636
+ ],
2637
+ "dtype": "float32",
2638
+ "format": "f32-to-bf16",
2639
+ "nbytes": 6144,
2640
+ "byteOffset": 31752192
2641
+ },
2642
+ {
2643
+ "name": "model.layers.24.post_attention_layernorm.weight",
2644
+ "shape": [
2645
+ 3072
2646
+ ],
2647
+ "dtype": "float32",
2648
+ "format": "f32-to-bf16",
2649
+ "nbytes": 6144,
2650
+ "byteOffset": 31758336
2651
+ },
2652
+ {
2653
+ "name": "model.layers.25.input_layernorm.weight",
2654
+ "shape": [
2655
+ 3072
2656
+ ],
2657
+ "dtype": "float32",
2658
+ "format": "f32-to-bf16",
2659
+ "nbytes": 6144,
2660
+ "byteOffset": 31764480
2661
+ },
2662
+ {
2663
+ "name": "model.layers.25.post_attention_layernorm.weight",
2664
+ "shape": [
2665
+ 3072
2666
+ ],
2667
+ "dtype": "float32",
2668
+ "format": "f32-to-bf16",
2669
+ "nbytes": 6144,
2670
+ "byteOffset": 31770624
2671
+ },
2672
+ {
2673
+ "name": "model.layers.26.input_layernorm.weight",
2674
+ "shape": [
2675
+ 3072
2676
+ ],
2677
+ "dtype": "float32",
2678
+ "format": "f32-to-bf16",
2679
+ "nbytes": 6144,
2680
+ "byteOffset": 31776768
2681
+ },
2682
+ {
2683
+ "name": "model.layers.26.post_attention_layernorm.weight",
2684
+ "shape": [
2685
+ 3072
2686
+ ],
2687
+ "dtype": "float32",
2688
+ "format": "f32-to-bf16",
2689
+ "nbytes": 6144,
2690
+ "byteOffset": 31782912
2691
+ },
2692
+ {
2693
+ "name": "model.layers.27.input_layernorm.weight",
2694
+ "shape": [
2695
+ 3072
2696
+ ],
2697
+ "dtype": "float32",
2698
+ "format": "f32-to-bf16",
2699
+ "nbytes": 6144,
2700
+ "byteOffset": 31789056
2701
+ },
2702
+ {
2703
+ "name": "model.layers.27.post_attention_layernorm.weight",
2704
+ "shape": [
2705
+ 3072
2706
+ ],
2707
+ "dtype": "float32",
2708
+ "format": "f32-to-bf16",
2709
+ "nbytes": 6144,
2710
+ "byteOffset": 31795200
2711
+ },
2712
+ {
2713
+ "name": "model.norm.weight",
2714
+ "shape": [
2715
+ 3072
2716
+ ],
2717
+ "dtype": "float32",
2718
+ "format": "f32-to-bf16",
2719
+ "nbytes": 6144,
2720
+ "byteOffset": 31801344
2721
+ }
2722
+ ],
2723
+ "md5sum": "8952ab985c888d6241e0e6204be82d7a"
2724
+ }
2725
+ ]
2726
+ }
params_shard_0.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c25862af666c12e7f95395e64d675b8294407af9b9df365c5f5a6b46d6ef938
3
+ size 788004864
params_shard_1.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb0966f2effb42ace9e6df82f7b4bc98ef89129394fc88e7369747f2658cae29
3
+ size 50331648
params_shard_10.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8fd52831d3c1ff296775305b51e9df7078e2ec653b596ca26e2cba081bcba7a5
3
+ size 31457280
params_shard_100.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:58dd2a492e91289446a6ae32fd6d9bcdfd139474e4323c630d9b66c490c22065
3
+ size 50331648
params_shard_101.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:407b4a2c93beb92dfe4db98bab7e1718b43075ad522f970706ebfb1870c41883
3
+ size 100663296
params_shard_102.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c3b059d8c716dc844b61dae7e9ec91aa091e401f57cd1b431905a8a4bbef659
3
+ size 31457280
params_shard_103.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b33146928d9a6a93fbf4d565a26af4126cf122f0573cda0eded29cbc12877c4b
3
+ size 18874368
params_shard_104.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bcd524a57a15757039e2cbcaf91e76874be77cd2546d88a92777c9734727074e
3
+ size 50331648
params_shard_105.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:57ac0574a30e1036df64491bdc61cc9bc930e61de1249b7b858502b50fb11798
3
+ size 100663296
params_shard_106.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7634bad6af805edca5b2ff5df43ee3c756c198b48c0c75627754a1d91c5f3e39
3
+ size 31457280
params_shard_107.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ab5a2421833961ea9700783395a2155fccfd6a79a1ea04a561583fcd6237369a
3
+ size 18874368
params_shard_108.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:27a7f8bea4a1b0c4061b5c2238baee7d3dd768e3e7957940b7fcdd1df1b294fb
3
+ size 50331648
params_shard_109.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ececa947a26a162d28865b16f8962ce0a2405b12ad7213b2fc7e3bd2718d30e2
3
+ size 100663296
params_shard_11.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7fbe7a49ad748e331e61be4021b760be1d1bc318640412e2c4ddc3b7551966e1
3
+ size 18874368
params_shard_110.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:95c712e7c8d1527a2ea9eaf1ff3c730e0b77cf8224ef608de81fefbc572986c7
3
+ size 31457280
params_shard_111.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d01d92102766e76b716a6a3093fd2126766e0591249cd98a1fea23290403d36
3
+ size 18874368
params_shard_112.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e810a74736e0afed559d0cd60ec64f80241e5910842b1b3e373fef0e0324cc32
3
+ size 31807488
params_shard_12.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf9881c3708d0a77e3c559fa6c70f392d2e9b99a9c948a67aec23e60597c88c4
3
+ size 50331648
params_shard_13.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:00c474352212e35714b3144dd3ce0bf9b8ab1371486bda283fcae6e5c00b4ce5
3
+ size 100663296
params_shard_14.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aec952dc6ab66caf4f9c2c8d98538c028ba8d216775da4a2f59743a631ee9b57
3
+ size 31457280
params_shard_15.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d6ee9d76a5c23ae382ade1405ebb4cd66f44d8e3e5ca7efee72381dabddfce1
3
+ size 18874368
params_shard_16.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:50b1d76892f4980c9674f9ed0c2444368c819b7e59a45bf3d78734a33b193365
3
+ size 50331648
params_shard_17.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bbc1a4f4d5fb6df882d7ad09620daf23bbecd6b01e9856ee9dc64d5fbf8519f6
3
+ size 100663296
params_shard_18.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6fd7319bd9b66409285094ce019f1c277713d87ed946a9c1cedfa1d75bc5936
3
+ size 31457280
params_shard_19.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:35a8e838a1314466fbdd76f364155810f3fc73bb7c7167ef9bd10d473f1eaf01
3
+ size 18874368
params_shard_2.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4812d3d10c7fb35a9aae5915cd6096dc09caf9db4eac50e8cb26d7c649cd2af5
3
+ size 100663296
params_shard_20.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:487e19af22ac85db161761e4fb593d3a19e46660356c74b23b090d392a93630a
3
+ size 50331648
params_shard_21.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:815cad9534ca34bc664a47a8bbec243a6ed19d77756ae9598da1af129129c224
3
+ size 100663296
params_shard_22.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:466b74fafae4d2de462bdaeba596d158c54f75dd64c02b5ff63092fe59ed2142
3
+ size 31457280
params_shard_23.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cd2e10141731348c44360a007f77ced034117a8bb17157c7cfe5d5f45552140d
3
+ size 18874368
params_shard_24.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:278e45c5be4f250ab6914d61da899b720a79fb3d9d338c6c4795a37d66938ddf
3
+ size 50331648
params_shard_25.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:568f00bcedaba748ad8cd7564dc928c5242202e797ef9c844b71275a06e55722
3
+ size 100663296
params_shard_26.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1ed7aeac39e16e451a51169610beeee9677c2d93b6fe579a8fcbdca78bcde9c5
3
+ size 31457280
params_shard_27.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5bfe5da9bb1c4779a74ce38dc2ac0bdfb59669b9e7e081cf3bac5be1f30d1221
3
+ size 18874368
params_shard_28.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:de3d9b14acfdc620f790eadc9183f216af03e312fbd5366f4c108969b99d374f
3
+ size 50331648
params_shard_29.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c440492c56290837dd152a78aceafbf128ca64eb7e52efb0a78e1d90a0e2c27f
3
+ size 100663296
params_shard_3.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2d36b6084bcb25809db28ff4ba0123ed890ec7439148b9b4bbb05ab9cf453d17
3
+ size 18874368
params_shard_30.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b27be5f0c624963ea5742af8dafa6916edd97740e5246b10ca325e96a9604aeb
3
+ size 31457280
params_shard_31.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a8a327277082da8427575a623ae6abd36ca277d2c8e580b5a349be8a3827e4e
3
+ size 18874368
params_shard_32.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e18e3ea31ee337242c56af8a37d2453cabec293988603e8748b7c592b38f55b
3
+ size 50331648
params_shard_33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3c75f2e8b8405f6955bd67fb8e93f25e66828531514ab0be2e41906b2fa1aed3
3
+ size 100663296
params_shard_34.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84fc71bd35ea14a2e2fbee1fb74736eee9640ffb0af67597d08141a2118d64fb
3
+ size 31457280
params_shard_35.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:78485b95d9a271f3d7460db24f6306a4245e321a43e391793e2243d2f98b728b
3
+ size 18874368
params_shard_36.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:167157286f9128595100b7a4fa30d3c7e7e45922d65652ee97effaaa08bb8bc3
3
+ size 50331648
params_shard_37.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e674582d6bae90bb1dc745e19c69e0fdd9f6e99dcf33230e999021f30f29962
3
+ size 100663296
params_shard_38.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:822f890e8ae3b33c81b8efdb719b45a5bab53eb818629c0ff08bd8859acf0227
3
+ size 31457280
params_shard_39.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f94b152d02d5f2a59714295e5c625206ddf7382798f8aefb52ffa82cbb91e3f5
3
+ size 18874368