riczhou commited on
Commit
ce7fdb0
1 Parent(s): 416a053

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
mlc-chat-config.json ADDED
@@ -0,0 +1,84 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "version": "0.1.0",
3
+ "model_type": "qwen2",
4
+ "quantization": "q0f16",
5
+ "model_config": {
6
+ "hidden_act": "silu",
7
+ "hidden_size": 1536,
8
+ "intermediate_size": 8960,
9
+ "num_attention_heads": 12,
10
+ "num_hidden_layers": 28,
11
+ "num_key_value_heads": 2,
12
+ "rms_norm_eps": 1e-06,
13
+ "rope_theta": 1000000.0,
14
+ "vocab_size": 151936,
15
+ "tie_word_embeddings": true,
16
+ "context_window_size": 32768,
17
+ "prefill_chunk_size": 2048,
18
+ "tensor_parallel_shards": 1,
19
+ "head_dim": 128,
20
+ "dtype": "float32",
21
+ "max_batch_size": 80
22
+ },
23
+ "vocab_size": 151936,
24
+ "context_window_size": 32768,
25
+ "sliding_window_size": -1,
26
+ "prefill_chunk_size": 2048,
27
+ "attention_sink_size": -1,
28
+ "tensor_parallel_shards": 1,
29
+ "pipeline_parallel_stages": 1,
30
+ "temperature": 0.7,
31
+ "presence_penalty": 0.0,
32
+ "frequency_penalty": 0.0,
33
+ "repetition_penalty": 1.1,
34
+ "top_p": 0.8,
35
+ "tokenizer_files": [
36
+ "tokenizer.json",
37
+ "vocab.json",
38
+ "merges.txt",
39
+ "tokenizer_config.json"
40
+ ],
41
+ "tokenizer_info": {
42
+ "token_postproc_method": "byte_level",
43
+ "prepend_space_in_encode": false,
44
+ "strip_space_in_decode": false
45
+ },
46
+ "conv_template": {
47
+ "name": "qwen2",
48
+ "system_template": "<|im_start|>system\n{system_message}<|im_end|>\n",
49
+ "system_message": "You are a helpful assistant.",
50
+ "system_prefix_token_ids": null,
51
+ "add_role_after_system_message": true,
52
+ "roles": {
53
+ "user": "<|im_start|>user",
54
+ "assistant": "<|im_start|>assistant"
55
+ },
56
+ "role_templates": {
57
+ "user": "{user_message}",
58
+ "assistant": "{assistant_message}",
59
+ "tool": "{tool_message}"
60
+ },
61
+ "messages": [],
62
+ "seps": [
63
+ "<|im_end|>\n"
64
+ ],
65
+ "role_content_sep": "\n",
66
+ "role_empty_sep": "\n",
67
+ "stop_str": [
68
+ "<|endoftext|>",
69
+ "<|im_end|>"
70
+ ],
71
+ "stop_token_ids": [
72
+ 151643,
73
+ 151645
74
+ ],
75
+ "function_string": "",
76
+ "use_function_calling": false
77
+ },
78
+ "pad_token_id": 151643,
79
+ "bos_token_id": 151643,
80
+ "eos_token_id": [
81
+ 151645,
82
+ 151643
83
+ ]
84
+ }
ndarray-cache.json ADDED
@@ -0,0 +1,2638 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "ParamSize": 198,
4
+ "ParamBytes": 3087428608.0,
5
+ "BitsPerParam": 16.0
6
+ },
7
+ "records": [
8
+ {
9
+ "dataPath": "params_shard_0.bin",
10
+ "format": "raw-shard",
11
+ "nbytes": 466747392,
12
+ "records": [
13
+ {
14
+ "name": "model.embed_tokens.weight",
15
+ "shape": [
16
+ 151936,
17
+ 1536
18
+ ],
19
+ "dtype": "float16",
20
+ "format": "f32-to-bf16",
21
+ "nbytes": 466747392,
22
+ "byteOffset": 0
23
+ }
24
+ ],
25
+ "md5sum": "fecf67195ad7c035b39a5b4b1c77facc"
26
+ },
27
+ {
28
+ "dataPath": "params_shard_1.bin",
29
+ "format": "raw-shard",
30
+ "nbytes": 55050240,
31
+ "records": [
32
+ {
33
+ "name": "model.layers.0.mlp.gate_up_proj.weight",
34
+ "shape": [
35
+ 17920,
36
+ 1536
37
+ ],
38
+ "dtype": "float16",
39
+ "format": "f32-to-bf16",
40
+ "nbytes": 55050240,
41
+ "byteOffset": 0
42
+ }
43
+ ],
44
+ "md5sum": "f01fe30dd4b66d59562cd7e350fe4688"
45
+ },
46
+ {
47
+ "dataPath": "params_shard_2.bin",
48
+ "format": "raw-shard",
49
+ "nbytes": 27535360,
50
+ "records": [
51
+ {
52
+ "name": "model.layers.0.input_layernorm.weight",
53
+ "shape": [
54
+ 1536
55
+ ],
56
+ "dtype": "float16",
57
+ "format": "f32-to-bf16",
58
+ "nbytes": 3072,
59
+ "byteOffset": 0
60
+ },
61
+ {
62
+ "name": "model.layers.0.mlp.down_proj.weight",
63
+ "shape": [
64
+ 1536,
65
+ 8960
66
+ ],
67
+ "dtype": "float16",
68
+ "format": "f32-to-bf16",
69
+ "nbytes": 27525120,
70
+ "byteOffset": 3072
71
+ },
72
+ {
73
+ "name": "model.layers.0.post_attention_layernorm.weight",
74
+ "shape": [
75
+ 1536
76
+ ],
77
+ "dtype": "float16",
78
+ "format": "f32-to-bf16",
79
+ "nbytes": 3072,
80
+ "byteOffset": 27528192
81
+ },
82
+ {
83
+ "name": "model.layers.0.self_attn.c_attn.bias",
84
+ "shape": [
85
+ 2048
86
+ ],
87
+ "dtype": "float16",
88
+ "format": "f32-to-bf16",
89
+ "nbytes": 4096,
90
+ "byteOffset": 27531264
91
+ }
92
+ ],
93
+ "md5sum": "53ba56fbef3f358710c88b7094762ab3"
94
+ },
95
+ {
96
+ "dataPath": "params_shard_3.bin",
97
+ "format": "raw-shard",
98
+ "nbytes": 27525120,
99
+ "records": [
100
+ {
101
+ "name": "model.layers.1.mlp.down_proj.weight",
102
+ "shape": [
103
+ 1536,
104
+ 8960
105
+ ],
106
+ "dtype": "float16",
107
+ "format": "f32-to-bf16",
108
+ "nbytes": 27525120,
109
+ "byteOffset": 0
110
+ }
111
+ ],
112
+ "md5sum": "87cd1576987924e553ea7ee2e7f19300"
113
+ },
114
+ {
115
+ "dataPath": "params_shard_4.bin",
116
+ "format": "raw-shard",
117
+ "nbytes": 55050240,
118
+ "records": [
119
+ {
120
+ "name": "model.layers.1.mlp.gate_up_proj.weight",
121
+ "shape": [
122
+ 17920,
123
+ 1536
124
+ ],
125
+ "dtype": "float16",
126
+ "format": "f32-to-bf16",
127
+ "nbytes": 55050240,
128
+ "byteOffset": 0
129
+ }
130
+ ],
131
+ "md5sum": "ddfbf20abada3bf31cc1294cbd2bbaa4"
132
+ },
133
+ {
134
+ "dataPath": "params_shard_5.bin",
135
+ "format": "raw-shard",
136
+ "nbytes": 27525120,
137
+ "records": [
138
+ {
139
+ "name": "model.layers.10.mlp.down_proj.weight",
140
+ "shape": [
141
+ 1536,
142
+ 8960
143
+ ],
144
+ "dtype": "float16",
145
+ "format": "f32-to-bf16",
146
+ "nbytes": 27525120,
147
+ "byteOffset": 0
148
+ }
149
+ ],
150
+ "md5sum": "79b51572a280ed240e3f211a3d4d2c58"
151
+ },
152
+ {
153
+ "dataPath": "params_shard_6.bin",
154
+ "format": "raw-shard",
155
+ "nbytes": 55050240,
156
+ "records": [
157
+ {
158
+ "name": "model.layers.10.mlp.gate_up_proj.weight",
159
+ "shape": [
160
+ 17920,
161
+ 1536
162
+ ],
163
+ "dtype": "float16",
164
+ "format": "f32-to-bf16",
165
+ "nbytes": 55050240,
166
+ "byteOffset": 0
167
+ }
168
+ ],
169
+ "md5sum": "2816556bc4da221cb6a750ee2ece563f"
170
+ },
171
+ {
172
+ "dataPath": "params_shard_7.bin",
173
+ "format": "raw-shard",
174
+ "nbytes": 27525120,
175
+ "records": [
176
+ {
177
+ "name": "model.layers.11.mlp.down_proj.weight",
178
+ "shape": [
179
+ 1536,
180
+ 8960
181
+ ],
182
+ "dtype": "float16",
183
+ "format": "f32-to-bf16",
184
+ "nbytes": 27525120,
185
+ "byteOffset": 0
186
+ }
187
+ ],
188
+ "md5sum": "15e2a91381e3ffcd35f6d2b11dc43693"
189
+ },
190
+ {
191
+ "dataPath": "params_shard_8.bin",
192
+ "format": "raw-shard",
193
+ "nbytes": 55050240,
194
+ "records": [
195
+ {
196
+ "name": "model.layers.11.mlp.gate_up_proj.weight",
197
+ "shape": [
198
+ 17920,
199
+ 1536
200
+ ],
201
+ "dtype": "float16",
202
+ "format": "f32-to-bf16",
203
+ "nbytes": 55050240,
204
+ "byteOffset": 0
205
+ }
206
+ ],
207
+ "md5sum": "246f21c55a225dcdcc5d87881bcb06b3"
208
+ },
209
+ {
210
+ "dataPath": "params_shard_9.bin",
211
+ "format": "raw-shard",
212
+ "nbytes": 33060864,
213
+ "records": [
214
+ {
215
+ "name": "model.layers.0.self_attn.c_attn.weight",
216
+ "shape": [
217
+ 2048,
218
+ 1536
219
+ ],
220
+ "dtype": "float16",
221
+ "format": "f32-to-bf16",
222
+ "nbytes": 6291456,
223
+ "byteOffset": 0
224
+ },
225
+ {
226
+ "name": "model.layers.0.self_attn.o_proj.weight",
227
+ "shape": [
228
+ 1536,
229
+ 1536
230
+ ],
231
+ "dtype": "float16",
232
+ "format": "f32-to-bf16",
233
+ "nbytes": 4718592,
234
+ "byteOffset": 6291456
235
+ },
236
+ {
237
+ "name": "model.layers.1.input_layernorm.weight",
238
+ "shape": [
239
+ 1536
240
+ ],
241
+ "dtype": "float16",
242
+ "format": "f32-to-bf16",
243
+ "nbytes": 3072,
244
+ "byteOffset": 11010048
245
+ },
246
+ {
247
+ "name": "model.layers.1.post_attention_layernorm.weight",
248
+ "shape": [
249
+ 1536
250
+ ],
251
+ "dtype": "float16",
252
+ "format": "f32-to-bf16",
253
+ "nbytes": 3072,
254
+ "byteOffset": 11013120
255
+ },
256
+ {
257
+ "name": "model.layers.1.self_attn.c_attn.bias",
258
+ "shape": [
259
+ 2048
260
+ ],
261
+ "dtype": "float16",
262
+ "format": "f32-to-bf16",
263
+ "nbytes": 4096,
264
+ "byteOffset": 11016192
265
+ },
266
+ {
267
+ "name": "model.layers.1.self_attn.c_attn.weight",
268
+ "shape": [
269
+ 2048,
270
+ 1536
271
+ ],
272
+ "dtype": "float16",
273
+ "format": "f32-to-bf16",
274
+ "nbytes": 6291456,
275
+ "byteOffset": 11020288
276
+ },
277
+ {
278
+ "name": "model.layers.1.self_attn.o_proj.weight",
279
+ "shape": [
280
+ 1536,
281
+ 1536
282
+ ],
283
+ "dtype": "float16",
284
+ "format": "f32-to-bf16",
285
+ "nbytes": 4718592,
286
+ "byteOffset": 17311744
287
+ },
288
+ {
289
+ "name": "model.layers.10.input_layernorm.weight",
290
+ "shape": [
291
+ 1536
292
+ ],
293
+ "dtype": "float16",
294
+ "format": "f32-to-bf16",
295
+ "nbytes": 3072,
296
+ "byteOffset": 22030336
297
+ },
298
+ {
299
+ "name": "model.layers.10.post_attention_layernorm.weight",
300
+ "shape": [
301
+ 1536
302
+ ],
303
+ "dtype": "float16",
304
+ "format": "f32-to-bf16",
305
+ "nbytes": 3072,
306
+ "byteOffset": 22033408
307
+ },
308
+ {
309
+ "name": "model.layers.10.self_attn.c_attn.bias",
310
+ "shape": [
311
+ 2048
312
+ ],
313
+ "dtype": "float16",
314
+ "format": "f32-to-bf16",
315
+ "nbytes": 4096,
316
+ "byteOffset": 22036480
317
+ },
318
+ {
319
+ "name": "model.layers.10.self_attn.c_attn.weight",
320
+ "shape": [
321
+ 2048,
322
+ 1536
323
+ ],
324
+ "dtype": "float16",
325
+ "format": "f32-to-bf16",
326
+ "nbytes": 6291456,
327
+ "byteOffset": 22040576
328
+ },
329
+ {
330
+ "name": "model.layers.10.self_attn.o_proj.weight",
331
+ "shape": [
332
+ 1536,
333
+ 1536
334
+ ],
335
+ "dtype": "float16",
336
+ "format": "f32-to-bf16",
337
+ "nbytes": 4718592,
338
+ "byteOffset": 28332032
339
+ },
340
+ {
341
+ "name": "model.layers.11.input_layernorm.weight",
342
+ "shape": [
343
+ 1536
344
+ ],
345
+ "dtype": "float16",
346
+ "format": "f32-to-bf16",
347
+ "nbytes": 3072,
348
+ "byteOffset": 33050624
349
+ },
350
+ {
351
+ "name": "model.layers.11.post_attention_layernorm.weight",
352
+ "shape": [
353
+ 1536
354
+ ],
355
+ "dtype": "float16",
356
+ "format": "f32-to-bf16",
357
+ "nbytes": 3072,
358
+ "byteOffset": 33053696
359
+ },
360
+ {
361
+ "name": "model.layers.11.self_attn.c_attn.bias",
362
+ "shape": [
363
+ 2048
364
+ ],
365
+ "dtype": "float16",
366
+ "format": "f32-to-bf16",
367
+ "nbytes": 4096,
368
+ "byteOffset": 33056768
369
+ }
370
+ ],
371
+ "md5sum": "9db0eff55267cb68ef2f15c9d8921f66"
372
+ },
373
+ {
374
+ "dataPath": "params_shard_10.bin",
375
+ "format": "raw-shard",
376
+ "nbytes": 27525120,
377
+ "records": [
378
+ {
379
+ "name": "model.layers.12.mlp.down_proj.weight",
380
+ "shape": [
381
+ 1536,
382
+ 8960
383
+ ],
384
+ "dtype": "float16",
385
+ "format": "f32-to-bf16",
386
+ "nbytes": 27525120,
387
+ "byteOffset": 0
388
+ }
389
+ ],
390
+ "md5sum": "389eb23cbcc8cdeb6cf0666c1c14716d"
391
+ },
392
+ {
393
+ "dataPath": "params_shard_11.bin",
394
+ "format": "raw-shard",
395
+ "nbytes": 55050240,
396
+ "records": [
397
+ {
398
+ "name": "model.layers.12.mlp.gate_up_proj.weight",
399
+ "shape": [
400
+ 17920,
401
+ 1536
402
+ ],
403
+ "dtype": "float16",
404
+ "format": "f32-to-bf16",
405
+ "nbytes": 55050240,
406
+ "byteOffset": 0
407
+ }
408
+ ],
409
+ "md5sum": "858ff0c7d7ef0468583b849c3c011533"
410
+ },
411
+ {
412
+ "dataPath": "params_shard_12.bin",
413
+ "format": "raw-shard",
414
+ "nbytes": 27525120,
415
+ "records": [
416
+ {
417
+ "name": "model.layers.13.mlp.down_proj.weight",
418
+ "shape": [
419
+ 1536,
420
+ 8960
421
+ ],
422
+ "dtype": "float16",
423
+ "format": "f32-to-bf16",
424
+ "nbytes": 27525120,
425
+ "byteOffset": 0
426
+ }
427
+ ],
428
+ "md5sum": "1e76825b651f7617ac67e943d9ceef3e"
429
+ },
430
+ {
431
+ "dataPath": "params_shard_13.bin",
432
+ "format": "raw-shard",
433
+ "nbytes": 55050240,
434
+ "records": [
435
+ {
436
+ "name": "model.layers.13.mlp.gate_up_proj.weight",
437
+ "shape": [
438
+ 17920,
439
+ 1536
440
+ ],
441
+ "dtype": "float16",
442
+ "format": "f32-to-bf16",
443
+ "nbytes": 55050240,
444
+ "byteOffset": 0
445
+ }
446
+ ],
447
+ "md5sum": "2362cf24665a73b622c3f392bbab76ee"
448
+ },
449
+ {
450
+ "dataPath": "params_shard_14.bin",
451
+ "format": "raw-shard",
452
+ "nbytes": 27525120,
453
+ "records": [
454
+ {
455
+ "name": "model.layers.14.mlp.down_proj.weight",
456
+ "shape": [
457
+ 1536,
458
+ 8960
459
+ ],
460
+ "dtype": "float16",
461
+ "format": "f32-to-bf16",
462
+ "nbytes": 27525120,
463
+ "byteOffset": 0
464
+ }
465
+ ],
466
+ "md5sum": "85e525d88ea7db0deb627ee1aa325d88"
467
+ },
468
+ {
469
+ "dataPath": "params_shard_15.bin",
470
+ "format": "raw-shard",
471
+ "nbytes": 55050240,
472
+ "records": [
473
+ {
474
+ "name": "model.layers.14.mlp.gate_up_proj.weight",
475
+ "shape": [
476
+ 17920,
477
+ 1536
478
+ ],
479
+ "dtype": "float16",
480
+ "format": "f32-to-bf16",
481
+ "nbytes": 55050240,
482
+ "byteOffset": 0
483
+ }
484
+ ],
485
+ "md5sum": "70a8d25f82c16b0c7a46f3148062fbc3"
486
+ },
487
+ {
488
+ "dataPath": "params_shard_16.bin",
489
+ "format": "raw-shard",
490
+ "nbytes": 33060864,
491
+ "records": [
492
+ {
493
+ "name": "model.layers.11.self_attn.c_attn.weight",
494
+ "shape": [
495
+ 2048,
496
+ 1536
497
+ ],
498
+ "dtype": "float16",
499
+ "format": "f32-to-bf16",
500
+ "nbytes": 6291456,
501
+ "byteOffset": 0
502
+ },
503
+ {
504
+ "name": "model.layers.11.self_attn.o_proj.weight",
505
+ "shape": [
506
+ 1536,
507
+ 1536
508
+ ],
509
+ "dtype": "float16",
510
+ "format": "f32-to-bf16",
511
+ "nbytes": 4718592,
512
+ "byteOffset": 6291456
513
+ },
514
+ {
515
+ "name": "model.layers.12.input_layernorm.weight",
516
+ "shape": [
517
+ 1536
518
+ ],
519
+ "dtype": "float16",
520
+ "format": "f32-to-bf16",
521
+ "nbytes": 3072,
522
+ "byteOffset": 11010048
523
+ },
524
+ {
525
+ "name": "model.layers.12.post_attention_layernorm.weight",
526
+ "shape": [
527
+ 1536
528
+ ],
529
+ "dtype": "float16",
530
+ "format": "f32-to-bf16",
531
+ "nbytes": 3072,
532
+ "byteOffset": 11013120
533
+ },
534
+ {
535
+ "name": "model.layers.12.self_attn.c_attn.bias",
536
+ "shape": [
537
+ 2048
538
+ ],
539
+ "dtype": "float16",
540
+ "format": "f32-to-bf16",
541
+ "nbytes": 4096,
542
+ "byteOffset": 11016192
543
+ },
544
+ {
545
+ "name": "model.layers.12.self_attn.c_attn.weight",
546
+ "shape": [
547
+ 2048,
548
+ 1536
549
+ ],
550
+ "dtype": "float16",
551
+ "format": "f32-to-bf16",
552
+ "nbytes": 6291456,
553
+ "byteOffset": 11020288
554
+ },
555
+ {
556
+ "name": "model.layers.12.self_attn.o_proj.weight",
557
+ "shape": [
558
+ 1536,
559
+ 1536
560
+ ],
561
+ "dtype": "float16",
562
+ "format": "f32-to-bf16",
563
+ "nbytes": 4718592,
564
+ "byteOffset": 17311744
565
+ },
566
+ {
567
+ "name": "model.layers.13.input_layernorm.weight",
568
+ "shape": [
569
+ 1536
570
+ ],
571
+ "dtype": "float16",
572
+ "format": "f32-to-bf16",
573
+ "nbytes": 3072,
574
+ "byteOffset": 22030336
575
+ },
576
+ {
577
+ "name": "model.layers.13.post_attention_layernorm.weight",
578
+ "shape": [
579
+ 1536
580
+ ],
581
+ "dtype": "float16",
582
+ "format": "f32-to-bf16",
583
+ "nbytes": 3072,
584
+ "byteOffset": 22033408
585
+ },
586
+ {
587
+ "name": "model.layers.13.self_attn.c_attn.bias",
588
+ "shape": [
589
+ 2048
590
+ ],
591
+ "dtype": "float16",
592
+ "format": "f32-to-bf16",
593
+ "nbytes": 4096,
594
+ "byteOffset": 22036480
595
+ },
596
+ {
597
+ "name": "model.layers.13.self_attn.c_attn.weight",
598
+ "shape": [
599
+ 2048,
600
+ 1536
601
+ ],
602
+ "dtype": "float16",
603
+ "format": "f32-to-bf16",
604
+ "nbytes": 6291456,
605
+ "byteOffset": 22040576
606
+ },
607
+ {
608
+ "name": "model.layers.13.self_attn.o_proj.weight",
609
+ "shape": [
610
+ 1536,
611
+ 1536
612
+ ],
613
+ "dtype": "float16",
614
+ "format": "f32-to-bf16",
615
+ "nbytes": 4718592,
616
+ "byteOffset": 28332032
617
+ },
618
+ {
619
+ "name": "model.layers.14.input_layernorm.weight",
620
+ "shape": [
621
+ 1536
622
+ ],
623
+ "dtype": "float16",
624
+ "format": "f32-to-bf16",
625
+ "nbytes": 3072,
626
+ "byteOffset": 33050624
627
+ },
628
+ {
629
+ "name": "model.layers.14.post_attention_layernorm.weight",
630
+ "shape": [
631
+ 1536
632
+ ],
633
+ "dtype": "float16",
634
+ "format": "f32-to-bf16",
635
+ "nbytes": 3072,
636
+ "byteOffset": 33053696
637
+ },
638
+ {
639
+ "name": "model.layers.14.self_attn.c_attn.bias",
640
+ "shape": [
641
+ 2048
642
+ ],
643
+ "dtype": "float16",
644
+ "format": "f32-to-bf16",
645
+ "nbytes": 4096,
646
+ "byteOffset": 33056768
647
+ }
648
+ ],
649
+ "md5sum": "fa92d7ef0417b9ce1906c462ab24ffea"
650
+ },
651
+ {
652
+ "dataPath": "params_shard_17.bin",
653
+ "format": "raw-shard",
654
+ "nbytes": 27525120,
655
+ "records": [
656
+ {
657
+ "name": "model.layers.15.mlp.down_proj.weight",
658
+ "shape": [
659
+ 1536,
660
+ 8960
661
+ ],
662
+ "dtype": "float16",
663
+ "format": "f32-to-bf16",
664
+ "nbytes": 27525120,
665
+ "byteOffset": 0
666
+ }
667
+ ],
668
+ "md5sum": "a881856ab5b7a115095f01b657d2a60b"
669
+ },
670
+ {
671
+ "dataPath": "params_shard_18.bin",
672
+ "format": "raw-shard",
673
+ "nbytes": 55050240,
674
+ "records": [
675
+ {
676
+ "name": "model.layers.15.mlp.gate_up_proj.weight",
677
+ "shape": [
678
+ 17920,
679
+ 1536
680
+ ],
681
+ "dtype": "float16",
682
+ "format": "f32-to-bf16",
683
+ "nbytes": 55050240,
684
+ "byteOffset": 0
685
+ }
686
+ ],
687
+ "md5sum": "d425ac3b47746585b8e097e41dfec7b2"
688
+ },
689
+ {
690
+ "dataPath": "params_shard_19.bin",
691
+ "format": "raw-shard",
692
+ "nbytes": 27525120,
693
+ "records": [
694
+ {
695
+ "name": "model.layers.16.mlp.down_proj.weight",
696
+ "shape": [
697
+ 1536,
698
+ 8960
699
+ ],
700
+ "dtype": "float16",
701
+ "format": "f32-to-bf16",
702
+ "nbytes": 27525120,
703
+ "byteOffset": 0
704
+ }
705
+ ],
706
+ "md5sum": "ca9ef1c6914a179e652d00d2064e50ef"
707
+ },
708
+ {
709
+ "dataPath": "params_shard_20.bin",
710
+ "format": "raw-shard",
711
+ "nbytes": 55050240,
712
+ "records": [
713
+ {
714
+ "name": "model.layers.16.mlp.gate_up_proj.weight",
715
+ "shape": [
716
+ 17920,
717
+ 1536
718
+ ],
719
+ "dtype": "float16",
720
+ "format": "f32-to-bf16",
721
+ "nbytes": 55050240,
722
+ "byteOffset": 0
723
+ }
724
+ ],
725
+ "md5sum": "7b991ce3e80b73c3e355753a6a230dea"
726
+ },
727
+ {
728
+ "dataPath": "params_shard_21.bin",
729
+ "format": "raw-shard",
730
+ "nbytes": 27525120,
731
+ "records": [
732
+ {
733
+ "name": "model.layers.17.mlp.down_proj.weight",
734
+ "shape": [
735
+ 1536,
736
+ 8960
737
+ ],
738
+ "dtype": "float16",
739
+ "format": "f32-to-bf16",
740
+ "nbytes": 27525120,
741
+ "byteOffset": 0
742
+ }
743
+ ],
744
+ "md5sum": "763efc5964c72d0f4dbba97e835a312f"
745
+ },
746
+ {
747
+ "dataPath": "params_shard_22.bin",
748
+ "format": "raw-shard",
749
+ "nbytes": 55050240,
750
+ "records": [
751
+ {
752
+ "name": "model.layers.17.mlp.gate_up_proj.weight",
753
+ "shape": [
754
+ 17920,
755
+ 1536
756
+ ],
757
+ "dtype": "float16",
758
+ "format": "f32-to-bf16",
759
+ "nbytes": 55050240,
760
+ "byteOffset": 0
761
+ }
762
+ ],
763
+ "md5sum": "a95da5ac965ef3acc6325c5835885dc5"
764
+ },
765
+ {
766
+ "dataPath": "params_shard_23.bin",
767
+ "format": "raw-shard",
768
+ "nbytes": 33060864,
769
+ "records": [
770
+ {
771
+ "name": "model.layers.14.self_attn.c_attn.weight",
772
+ "shape": [
773
+ 2048,
774
+ 1536
775
+ ],
776
+ "dtype": "float16",
777
+ "format": "f32-to-bf16",
778
+ "nbytes": 6291456,
779
+ "byteOffset": 0
780
+ },
781
+ {
782
+ "name": "model.layers.14.self_attn.o_proj.weight",
783
+ "shape": [
784
+ 1536,
785
+ 1536
786
+ ],
787
+ "dtype": "float16",
788
+ "format": "f32-to-bf16",
789
+ "nbytes": 4718592,
790
+ "byteOffset": 6291456
791
+ },
792
+ {
793
+ "name": "model.layers.15.input_layernorm.weight",
794
+ "shape": [
795
+ 1536
796
+ ],
797
+ "dtype": "float16",
798
+ "format": "f32-to-bf16",
799
+ "nbytes": 3072,
800
+ "byteOffset": 11010048
801
+ },
802
+ {
803
+ "name": "model.layers.15.post_attention_layernorm.weight",
804
+ "shape": [
805
+ 1536
806
+ ],
807
+ "dtype": "float16",
808
+ "format": "f32-to-bf16",
809
+ "nbytes": 3072,
810
+ "byteOffset": 11013120
811
+ },
812
+ {
813
+ "name": "model.layers.15.self_attn.c_attn.bias",
814
+ "shape": [
815
+ 2048
816
+ ],
817
+ "dtype": "float16",
818
+ "format": "f32-to-bf16",
819
+ "nbytes": 4096,
820
+ "byteOffset": 11016192
821
+ },
822
+ {
823
+ "name": "model.layers.15.self_attn.c_attn.weight",
824
+ "shape": [
825
+ 2048,
826
+ 1536
827
+ ],
828
+ "dtype": "float16",
829
+ "format": "f32-to-bf16",
830
+ "nbytes": 6291456,
831
+ "byteOffset": 11020288
832
+ },
833
+ {
834
+ "name": "model.layers.15.self_attn.o_proj.weight",
835
+ "shape": [
836
+ 1536,
837
+ 1536
838
+ ],
839
+ "dtype": "float16",
840
+ "format": "f32-to-bf16",
841
+ "nbytes": 4718592,
842
+ "byteOffset": 17311744
843
+ },
844
+ {
845
+ "name": "model.layers.16.input_layernorm.weight",
846
+ "shape": [
847
+ 1536
848
+ ],
849
+ "dtype": "float16",
850
+ "format": "f32-to-bf16",
851
+ "nbytes": 3072,
852
+ "byteOffset": 22030336
853
+ },
854
+ {
855
+ "name": "model.layers.16.post_attention_layernorm.weight",
856
+ "shape": [
857
+ 1536
858
+ ],
859
+ "dtype": "float16",
860
+ "format": "f32-to-bf16",
861
+ "nbytes": 3072,
862
+ "byteOffset": 22033408
863
+ },
864
+ {
865
+ "name": "model.layers.16.self_attn.c_attn.bias",
866
+ "shape": [
867
+ 2048
868
+ ],
869
+ "dtype": "float16",
870
+ "format": "f32-to-bf16",
871
+ "nbytes": 4096,
872
+ "byteOffset": 22036480
873
+ },
874
+ {
875
+ "name": "model.layers.16.self_attn.c_attn.weight",
876
+ "shape": [
877
+ 2048,
878
+ 1536
879
+ ],
880
+ "dtype": "float16",
881
+ "format": "f32-to-bf16",
882
+ "nbytes": 6291456,
883
+ "byteOffset": 22040576
884
+ },
885
+ {
886
+ "name": "model.layers.16.self_attn.o_proj.weight",
887
+ "shape": [
888
+ 1536,
889
+ 1536
890
+ ],
891
+ "dtype": "float16",
892
+ "format": "f32-to-bf16",
893
+ "nbytes": 4718592,
894
+ "byteOffset": 28332032
895
+ },
896
+ {
897
+ "name": "model.layers.17.input_layernorm.weight",
898
+ "shape": [
899
+ 1536
900
+ ],
901
+ "dtype": "float16",
902
+ "format": "f32-to-bf16",
903
+ "nbytes": 3072,
904
+ "byteOffset": 33050624
905
+ },
906
+ {
907
+ "name": "model.layers.17.post_attention_layernorm.weight",
908
+ "shape": [
909
+ 1536
910
+ ],
911
+ "dtype": "float16",
912
+ "format": "f32-to-bf16",
913
+ "nbytes": 3072,
914
+ "byteOffset": 33053696
915
+ },
916
+ {
917
+ "name": "model.layers.17.self_attn.c_attn.bias",
918
+ "shape": [
919
+ 2048
920
+ ],
921
+ "dtype": "float16",
922
+ "format": "f32-to-bf16",
923
+ "nbytes": 4096,
924
+ "byteOffset": 33056768
925
+ }
926
+ ],
927
+ "md5sum": "5f942015a57b4111606653701b142a7f"
928
+ },
929
+ {
930
+ "dataPath": "params_shard_24.bin",
931
+ "format": "raw-shard",
932
+ "nbytes": 27525120,
933
+ "records": [
934
+ {
935
+ "name": "model.layers.18.mlp.down_proj.weight",
936
+ "shape": [
937
+ 1536,
938
+ 8960
939
+ ],
940
+ "dtype": "float16",
941
+ "format": "f32-to-bf16",
942
+ "nbytes": 27525120,
943
+ "byteOffset": 0
944
+ }
945
+ ],
946
+ "md5sum": "13680242c8c7ab440e5b7b82282a8290"
947
+ },
948
+ {
949
+ "dataPath": "params_shard_25.bin",
950
+ "format": "raw-shard",
951
+ "nbytes": 55050240,
952
+ "records": [
953
+ {
954
+ "name": "model.layers.18.mlp.gate_up_proj.weight",
955
+ "shape": [
956
+ 17920,
957
+ 1536
958
+ ],
959
+ "dtype": "float16",
960
+ "format": "f32-to-bf16",
961
+ "nbytes": 55050240,
962
+ "byteOffset": 0
963
+ }
964
+ ],
965
+ "md5sum": "7b83dd8b4e0735e958ed5c8bb941ba7e"
966
+ },
967
+ {
968
+ "dataPath": "params_shard_26.bin",
969
+ "format": "raw-shard",
970
+ "nbytes": 27525120,
971
+ "records": [
972
+ {
973
+ "name": "model.layers.19.mlp.down_proj.weight",
974
+ "shape": [
975
+ 1536,
976
+ 8960
977
+ ],
978
+ "dtype": "float16",
979
+ "format": "f32-to-bf16",
980
+ "nbytes": 27525120,
981
+ "byteOffset": 0
982
+ }
983
+ ],
984
+ "md5sum": "932245788f69d73fc5e4c8180f9a4044"
985
+ },
986
+ {
987
+ "dataPath": "params_shard_27.bin",
988
+ "format": "raw-shard",
989
+ "nbytes": 55050240,
990
+ "records": [
991
+ {
992
+ "name": "model.layers.19.mlp.gate_up_proj.weight",
993
+ "shape": [
994
+ 17920,
995
+ 1536
996
+ ],
997
+ "dtype": "float16",
998
+ "format": "f32-to-bf16",
999
+ "nbytes": 55050240,
1000
+ "byteOffset": 0
1001
+ }
1002
+ ],
1003
+ "md5sum": "2c62b527c4cbb5b01b90ef0e9b49c822"
1004
+ },
1005
+ {
1006
+ "dataPath": "params_shard_28.bin",
1007
+ "format": "raw-shard",
1008
+ "nbytes": 27525120,
1009
+ "records": [
1010
+ {
1011
+ "name": "model.layers.2.mlp.down_proj.weight",
1012
+ "shape": [
1013
+ 1536,
1014
+ 8960
1015
+ ],
1016
+ "dtype": "float16",
1017
+ "format": "f32-to-bf16",
1018
+ "nbytes": 27525120,
1019
+ "byteOffset": 0
1020
+ }
1021
+ ],
1022
+ "md5sum": "cb6357e5a698342b07c422e8fbaf9c36"
1023
+ },
1024
+ {
1025
+ "dataPath": "params_shard_29.bin",
1026
+ "format": "raw-shard",
1027
+ "nbytes": 55050240,
1028
+ "records": [
1029
+ {
1030
+ "name": "model.layers.2.mlp.gate_up_proj.weight",
1031
+ "shape": [
1032
+ 17920,
1033
+ 1536
1034
+ ],
1035
+ "dtype": "float16",
1036
+ "format": "f32-to-bf16",
1037
+ "nbytes": 55050240,
1038
+ "byteOffset": 0
1039
+ }
1040
+ ],
1041
+ "md5sum": "00d73f236aea765533c6d52f87d7af97"
1042
+ },
1043
+ {
1044
+ "dataPath": "params_shard_30.bin",
1045
+ "format": "raw-shard",
1046
+ "nbytes": 33060864,
1047
+ "records": [
1048
+ {
1049
+ "name": "model.layers.17.self_attn.c_attn.weight",
1050
+ "shape": [
1051
+ 2048,
1052
+ 1536
1053
+ ],
1054
+ "dtype": "float16",
1055
+ "format": "f32-to-bf16",
1056
+ "nbytes": 6291456,
1057
+ "byteOffset": 0
1058
+ },
1059
+ {
1060
+ "name": "model.layers.17.self_attn.o_proj.weight",
1061
+ "shape": [
1062
+ 1536,
1063
+ 1536
1064
+ ],
1065
+ "dtype": "float16",
1066
+ "format": "f32-to-bf16",
1067
+ "nbytes": 4718592,
1068
+ "byteOffset": 6291456
1069
+ },
1070
+ {
1071
+ "name": "model.layers.18.input_layernorm.weight",
1072
+ "shape": [
1073
+ 1536
1074
+ ],
1075
+ "dtype": "float16",
1076
+ "format": "f32-to-bf16",
1077
+ "nbytes": 3072,
1078
+ "byteOffset": 11010048
1079
+ },
1080
+ {
1081
+ "name": "model.layers.18.post_attention_layernorm.weight",
1082
+ "shape": [
1083
+ 1536
1084
+ ],
1085
+ "dtype": "float16",
1086
+ "format": "f32-to-bf16",
1087
+ "nbytes": 3072,
1088
+ "byteOffset": 11013120
1089
+ },
1090
+ {
1091
+ "name": "model.layers.18.self_attn.c_attn.bias",
1092
+ "shape": [
1093
+ 2048
1094
+ ],
1095
+ "dtype": "float16",
1096
+ "format": "f32-to-bf16",
1097
+ "nbytes": 4096,
1098
+ "byteOffset": 11016192
1099
+ },
1100
+ {
1101
+ "name": "model.layers.18.self_attn.c_attn.weight",
1102
+ "shape": [
1103
+ 2048,
1104
+ 1536
1105
+ ],
1106
+ "dtype": "float16",
1107
+ "format": "f32-to-bf16",
1108
+ "nbytes": 6291456,
1109
+ "byteOffset": 11020288
1110
+ },
1111
+ {
1112
+ "name": "model.layers.18.self_attn.o_proj.weight",
1113
+ "shape": [
1114
+ 1536,
1115
+ 1536
1116
+ ],
1117
+ "dtype": "float16",
1118
+ "format": "f32-to-bf16",
1119
+ "nbytes": 4718592,
1120
+ "byteOffset": 17311744
1121
+ },
1122
+ {
1123
+ "name": "model.layers.19.input_layernorm.weight",
1124
+ "shape": [
1125
+ 1536
1126
+ ],
1127
+ "dtype": "float16",
1128
+ "format": "f32-to-bf16",
1129
+ "nbytes": 3072,
1130
+ "byteOffset": 22030336
1131
+ },
1132
+ {
1133
+ "name": "model.layers.19.post_attention_layernorm.weight",
1134
+ "shape": [
1135
+ 1536
1136
+ ],
1137
+ "dtype": "float16",
1138
+ "format": "f32-to-bf16",
1139
+ "nbytes": 3072,
1140
+ "byteOffset": 22033408
1141
+ },
1142
+ {
1143
+ "name": "model.layers.19.self_attn.c_attn.bias",
1144
+ "shape": [
1145
+ 2048
1146
+ ],
1147
+ "dtype": "float16",
1148
+ "format": "f32-to-bf16",
1149
+ "nbytes": 4096,
1150
+ "byteOffset": 22036480
1151
+ },
1152
+ {
1153
+ "name": "model.layers.19.self_attn.c_attn.weight",
1154
+ "shape": [
1155
+ 2048,
1156
+ 1536
1157
+ ],
1158
+ "dtype": "float16",
1159
+ "format": "f32-to-bf16",
1160
+ "nbytes": 6291456,
1161
+ "byteOffset": 22040576
1162
+ },
1163
+ {
1164
+ "name": "model.layers.19.self_attn.o_proj.weight",
1165
+ "shape": [
1166
+ 1536,
1167
+ 1536
1168
+ ],
1169
+ "dtype": "float16",
1170
+ "format": "f32-to-bf16",
1171
+ "nbytes": 4718592,
1172
+ "byteOffset": 28332032
1173
+ },
1174
+ {
1175
+ "name": "model.layers.2.input_layernorm.weight",
1176
+ "shape": [
1177
+ 1536
1178
+ ],
1179
+ "dtype": "float16",
1180
+ "format": "f32-to-bf16",
1181
+ "nbytes": 3072,
1182
+ "byteOffset": 33050624
1183
+ },
1184
+ {
1185
+ "name": "model.layers.2.post_attention_layernorm.weight",
1186
+ "shape": [
1187
+ 1536
1188
+ ],
1189
+ "dtype": "float16",
1190
+ "format": "f32-to-bf16",
1191
+ "nbytes": 3072,
1192
+ "byteOffset": 33053696
1193
+ },
1194
+ {
1195
+ "name": "model.layers.2.self_attn.c_attn.bias",
1196
+ "shape": [
1197
+ 2048
1198
+ ],
1199
+ "dtype": "float16",
1200
+ "format": "f32-to-bf16",
1201
+ "nbytes": 4096,
1202
+ "byteOffset": 33056768
1203
+ }
1204
+ ],
1205
+ "md5sum": "7fd4e25e6520542170a8eb83dd34b808"
1206
+ },
1207
+ {
1208
+ "dataPath": "params_shard_31.bin",
1209
+ "format": "raw-shard",
1210
+ "nbytes": 27525120,
1211
+ "records": [
1212
+ {
1213
+ "name": "model.layers.20.mlp.down_proj.weight",
1214
+ "shape": [
1215
+ 1536,
1216
+ 8960
1217
+ ],
1218
+ "dtype": "float16",
1219
+ "format": "f32-to-bf16",
1220
+ "nbytes": 27525120,
1221
+ "byteOffset": 0
1222
+ }
1223
+ ],
1224
+ "md5sum": "0370a3b61e1104aade32736eed768f20"
1225
+ },
1226
+ {
1227
+ "dataPath": "params_shard_32.bin",
1228
+ "format": "raw-shard",
1229
+ "nbytes": 55050240,
1230
+ "records": [
1231
+ {
1232
+ "name": "model.layers.20.mlp.gate_up_proj.weight",
1233
+ "shape": [
1234
+ 17920,
1235
+ 1536
1236
+ ],
1237
+ "dtype": "float16",
1238
+ "format": "f32-to-bf16",
1239
+ "nbytes": 55050240,
1240
+ "byteOffset": 0
1241
+ }
1242
+ ],
1243
+ "md5sum": "d8848dd503a0c9a67908fdf6138be7d2"
1244
+ },
1245
+ {
1246
+ "dataPath": "params_shard_33.bin",
1247
+ "format": "raw-shard",
1248
+ "nbytes": 27525120,
1249
+ "records": [
1250
+ {
1251
+ "name": "model.layers.21.mlp.down_proj.weight",
1252
+ "shape": [
1253
+ 1536,
1254
+ 8960
1255
+ ],
1256
+ "dtype": "float16",
1257
+ "format": "f32-to-bf16",
1258
+ "nbytes": 27525120,
1259
+ "byteOffset": 0
1260
+ }
1261
+ ],
1262
+ "md5sum": "06075d5146ba69e8df5531112f0f9e1b"
1263
+ },
1264
+ {
1265
+ "dataPath": "params_shard_34.bin",
1266
+ "format": "raw-shard",
1267
+ "nbytes": 55050240,
1268
+ "records": [
1269
+ {
1270
+ "name": "model.layers.21.mlp.gate_up_proj.weight",
1271
+ "shape": [
1272
+ 17920,
1273
+ 1536
1274
+ ],
1275
+ "dtype": "float16",
1276
+ "format": "f32-to-bf16",
1277
+ "nbytes": 55050240,
1278
+ "byteOffset": 0
1279
+ }
1280
+ ],
1281
+ "md5sum": "3d3ef7a857a7c6fe81f5af14755b9ab4"
1282
+ },
1283
+ {
1284
+ "dataPath": "params_shard_35.bin",
1285
+ "format": "raw-shard",
1286
+ "nbytes": 27525120,
1287
+ "records": [
1288
+ {
1289
+ "name": "model.layers.22.mlp.down_proj.weight",
1290
+ "shape": [
1291
+ 1536,
1292
+ 8960
1293
+ ],
1294
+ "dtype": "float16",
1295
+ "format": "f32-to-bf16",
1296
+ "nbytes": 27525120,
1297
+ "byteOffset": 0
1298
+ }
1299
+ ],
1300
+ "md5sum": "83b82ceda7135499fc6b9c23be8ac81d"
1301
+ },
1302
+ {
1303
+ "dataPath": "params_shard_36.bin",
1304
+ "format": "raw-shard",
1305
+ "nbytes": 55050240,
1306
+ "records": [
1307
+ {
1308
+ "name": "model.layers.22.mlp.gate_up_proj.weight",
1309
+ "shape": [
1310
+ 17920,
1311
+ 1536
1312
+ ],
1313
+ "dtype": "float16",
1314
+ "format": "f32-to-bf16",
1315
+ "nbytes": 55050240,
1316
+ "byteOffset": 0
1317
+ }
1318
+ ],
1319
+ "md5sum": "0b9d95eebe1ead5366d33d19b6053dfa"
1320
+ },
1321
+ {
1322
+ "dataPath": "params_shard_37.bin",
1323
+ "format": "raw-shard",
1324
+ "nbytes": 33060864,
1325
+ "records": [
1326
+ {
1327
+ "name": "model.layers.2.self_attn.c_attn.weight",
1328
+ "shape": [
1329
+ 2048,
1330
+ 1536
1331
+ ],
1332
+ "dtype": "float16",
1333
+ "format": "f32-to-bf16",
1334
+ "nbytes": 6291456,
1335
+ "byteOffset": 0
1336
+ },
1337
+ {
1338
+ "name": "model.layers.2.self_attn.o_proj.weight",
1339
+ "shape": [
1340
+ 1536,
1341
+ 1536
1342
+ ],
1343
+ "dtype": "float16",
1344
+ "format": "f32-to-bf16",
1345
+ "nbytes": 4718592,
1346
+ "byteOffset": 6291456
1347
+ },
1348
+ {
1349
+ "name": "model.layers.20.input_layernorm.weight",
1350
+ "shape": [
1351
+ 1536
1352
+ ],
1353
+ "dtype": "float16",
1354
+ "format": "f32-to-bf16",
1355
+ "nbytes": 3072,
1356
+ "byteOffset": 11010048
1357
+ },
1358
+ {
1359
+ "name": "model.layers.20.post_attention_layernorm.weight",
1360
+ "shape": [
1361
+ 1536
1362
+ ],
1363
+ "dtype": "float16",
1364
+ "format": "f32-to-bf16",
1365
+ "nbytes": 3072,
1366
+ "byteOffset": 11013120
1367
+ },
1368
+ {
1369
+ "name": "model.layers.20.self_attn.c_attn.bias",
1370
+ "shape": [
1371
+ 2048
1372
+ ],
1373
+ "dtype": "float16",
1374
+ "format": "f32-to-bf16",
1375
+ "nbytes": 4096,
1376
+ "byteOffset": 11016192
1377
+ },
1378
+ {
1379
+ "name": "model.layers.20.self_attn.c_attn.weight",
1380
+ "shape": [
1381
+ 2048,
1382
+ 1536
1383
+ ],
1384
+ "dtype": "float16",
1385
+ "format": "f32-to-bf16",
1386
+ "nbytes": 6291456,
1387
+ "byteOffset": 11020288
1388
+ },
1389
+ {
1390
+ "name": "model.layers.20.self_attn.o_proj.weight",
1391
+ "shape": [
1392
+ 1536,
1393
+ 1536
1394
+ ],
1395
+ "dtype": "float16",
1396
+ "format": "f32-to-bf16",
1397
+ "nbytes": 4718592,
1398
+ "byteOffset": 17311744
1399
+ },
1400
+ {
1401
+ "name": "model.layers.21.input_layernorm.weight",
1402
+ "shape": [
1403
+ 1536
1404
+ ],
1405
+ "dtype": "float16",
1406
+ "format": "f32-to-bf16",
1407
+ "nbytes": 3072,
1408
+ "byteOffset": 22030336
1409
+ },
1410
+ {
1411
+ "name": "model.layers.21.post_attention_layernorm.weight",
1412
+ "shape": [
1413
+ 1536
1414
+ ],
1415
+ "dtype": "float16",
1416
+ "format": "f32-to-bf16",
1417
+ "nbytes": 3072,
1418
+ "byteOffset": 22033408
1419
+ },
1420
+ {
1421
+ "name": "model.layers.21.self_attn.c_attn.bias",
1422
+ "shape": [
1423
+ 2048
1424
+ ],
1425
+ "dtype": "float16",
1426
+ "format": "f32-to-bf16",
1427
+ "nbytes": 4096,
1428
+ "byteOffset": 22036480
1429
+ },
1430
+ {
1431
+ "name": "model.layers.21.self_attn.c_attn.weight",
1432
+ "shape": [
1433
+ 2048,
1434
+ 1536
1435
+ ],
1436
+ "dtype": "float16",
1437
+ "format": "f32-to-bf16",
1438
+ "nbytes": 6291456,
1439
+ "byteOffset": 22040576
1440
+ },
1441
+ {
1442
+ "name": "model.layers.21.self_attn.o_proj.weight",
1443
+ "shape": [
1444
+ 1536,
1445
+ 1536
1446
+ ],
1447
+ "dtype": "float16",
1448
+ "format": "f32-to-bf16",
1449
+ "nbytes": 4718592,
1450
+ "byteOffset": 28332032
1451
+ },
1452
+ {
1453
+ "name": "model.layers.22.input_layernorm.weight",
1454
+ "shape": [
1455
+ 1536
1456
+ ],
1457
+ "dtype": "float16",
1458
+ "format": "f32-to-bf16",
1459
+ "nbytes": 3072,
1460
+ "byteOffset": 33050624
1461
+ },
1462
+ {
1463
+ "name": "model.layers.22.post_attention_layernorm.weight",
1464
+ "shape": [
1465
+ 1536
1466
+ ],
1467
+ "dtype": "float16",
1468
+ "format": "f32-to-bf16",
1469
+ "nbytes": 3072,
1470
+ "byteOffset": 33053696
1471
+ },
1472
+ {
1473
+ "name": "model.layers.22.self_attn.c_attn.bias",
1474
+ "shape": [
1475
+ 2048
1476
+ ],
1477
+ "dtype": "float16",
1478
+ "format": "f32-to-bf16",
1479
+ "nbytes": 4096,
1480
+ "byteOffset": 33056768
1481
+ }
1482
+ ],
1483
+ "md5sum": "03bef970c2cedd8d2eac438675f1df1b"
1484
+ },
1485
+ {
1486
+ "dataPath": "params_shard_38.bin",
1487
+ "format": "raw-shard",
1488
+ "nbytes": 27525120,
1489
+ "records": [
1490
+ {
1491
+ "name": "model.layers.23.mlp.down_proj.weight",
1492
+ "shape": [
1493
+ 1536,
1494
+ 8960
1495
+ ],
1496
+ "dtype": "float16",
1497
+ "format": "f32-to-bf16",
1498
+ "nbytes": 27525120,
1499
+ "byteOffset": 0
1500
+ }
1501
+ ],
1502
+ "md5sum": "08d613a9ff790412962fd12134ab3a1e"
1503
+ },
1504
+ {
1505
+ "dataPath": "params_shard_39.bin",
1506
+ "format": "raw-shard",
1507
+ "nbytes": 55050240,
1508
+ "records": [
1509
+ {
1510
+ "name": "model.layers.23.mlp.gate_up_proj.weight",
1511
+ "shape": [
1512
+ 17920,
1513
+ 1536
1514
+ ],
1515
+ "dtype": "float16",
1516
+ "format": "f32-to-bf16",
1517
+ "nbytes": 55050240,
1518
+ "byteOffset": 0
1519
+ }
1520
+ ],
1521
+ "md5sum": "c1620922ea2373b51b87463527663182"
1522
+ },
1523
+ {
1524
+ "dataPath": "params_shard_40.bin",
1525
+ "format": "raw-shard",
1526
+ "nbytes": 27525120,
1527
+ "records": [
1528
+ {
1529
+ "name": "model.layers.24.mlp.down_proj.weight",
1530
+ "shape": [
1531
+ 1536,
1532
+ 8960
1533
+ ],
1534
+ "dtype": "float16",
1535
+ "format": "f32-to-bf16",
1536
+ "nbytes": 27525120,
1537
+ "byteOffset": 0
1538
+ }
1539
+ ],
1540
+ "md5sum": "e61a68e14bce574f00d6b15f6f1829ff"
1541
+ },
1542
+ {
1543
+ "dataPath": "params_shard_41.bin",
1544
+ "format": "raw-shard",
1545
+ "nbytes": 55050240,
1546
+ "records": [
1547
+ {
1548
+ "name": "model.layers.24.mlp.gate_up_proj.weight",
1549
+ "shape": [
1550
+ 17920,
1551
+ 1536
1552
+ ],
1553
+ "dtype": "float16",
1554
+ "format": "f32-to-bf16",
1555
+ "nbytes": 55050240,
1556
+ "byteOffset": 0
1557
+ }
1558
+ ],
1559
+ "md5sum": "740fed33f6bf3df4c44b826d3f957144"
1560
+ },
1561
+ {
1562
+ "dataPath": "params_shard_42.bin",
1563
+ "format": "raw-shard",
1564
+ "nbytes": 27525120,
1565
+ "records": [
1566
+ {
1567
+ "name": "model.layers.25.mlp.down_proj.weight",
1568
+ "shape": [
1569
+ 1536,
1570
+ 8960
1571
+ ],
1572
+ "dtype": "float16",
1573
+ "format": "f32-to-bf16",
1574
+ "nbytes": 27525120,
1575
+ "byteOffset": 0
1576
+ }
1577
+ ],
1578
+ "md5sum": "c688dbbdc6f8edcf61a10ef62c003802"
1579
+ },
1580
+ {
1581
+ "dataPath": "params_shard_43.bin",
1582
+ "format": "raw-shard",
1583
+ "nbytes": 55050240,
1584
+ "records": [
1585
+ {
1586
+ "name": "model.layers.25.mlp.gate_up_proj.weight",
1587
+ "shape": [
1588
+ 17920,
1589
+ 1536
1590
+ ],
1591
+ "dtype": "float16",
1592
+ "format": "f32-to-bf16",
1593
+ "nbytes": 55050240,
1594
+ "byteOffset": 0
1595
+ }
1596
+ ],
1597
+ "md5sum": "88def80b770c79315621af9521be59d8"
1598
+ },
1599
+ {
1600
+ "dataPath": "params_shard_44.bin",
1601
+ "format": "raw-shard",
1602
+ "nbytes": 33060864,
1603
+ "records": [
1604
+ {
1605
+ "name": "model.layers.22.self_attn.c_attn.weight",
1606
+ "shape": [
1607
+ 2048,
1608
+ 1536
1609
+ ],
1610
+ "dtype": "float16",
1611
+ "format": "f32-to-bf16",
1612
+ "nbytes": 6291456,
1613
+ "byteOffset": 0
1614
+ },
1615
+ {
1616
+ "name": "model.layers.22.self_attn.o_proj.weight",
1617
+ "shape": [
1618
+ 1536,
1619
+ 1536
1620
+ ],
1621
+ "dtype": "float16",
1622
+ "format": "f32-to-bf16",
1623
+ "nbytes": 4718592,
1624
+ "byteOffset": 6291456
1625
+ },
1626
+ {
1627
+ "name": "model.layers.23.input_layernorm.weight",
1628
+ "shape": [
1629
+ 1536
1630
+ ],
1631
+ "dtype": "float16",
1632
+ "format": "f32-to-bf16",
1633
+ "nbytes": 3072,
1634
+ "byteOffset": 11010048
1635
+ },
1636
+ {
1637
+ "name": "model.layers.23.post_attention_layernorm.weight",
1638
+ "shape": [
1639
+ 1536
1640
+ ],
1641
+ "dtype": "float16",
1642
+ "format": "f32-to-bf16",
1643
+ "nbytes": 3072,
1644
+ "byteOffset": 11013120
1645
+ },
1646
+ {
1647
+ "name": "model.layers.23.self_attn.c_attn.bias",
1648
+ "shape": [
1649
+ 2048
1650
+ ],
1651
+ "dtype": "float16",
1652
+ "format": "f32-to-bf16",
1653
+ "nbytes": 4096,
1654
+ "byteOffset": 11016192
1655
+ },
1656
+ {
1657
+ "name": "model.layers.23.self_attn.c_attn.weight",
1658
+ "shape": [
1659
+ 2048,
1660
+ 1536
1661
+ ],
1662
+ "dtype": "float16",
1663
+ "format": "f32-to-bf16",
1664
+ "nbytes": 6291456,
1665
+ "byteOffset": 11020288
1666
+ },
1667
+ {
1668
+ "name": "model.layers.23.self_attn.o_proj.weight",
1669
+ "shape": [
1670
+ 1536,
1671
+ 1536
1672
+ ],
1673
+ "dtype": "float16",
1674
+ "format": "f32-to-bf16",
1675
+ "nbytes": 4718592,
1676
+ "byteOffset": 17311744
1677
+ },
1678
+ {
1679
+ "name": "model.layers.24.input_layernorm.weight",
1680
+ "shape": [
1681
+ 1536
1682
+ ],
1683
+ "dtype": "float16",
1684
+ "format": "f32-to-bf16",
1685
+ "nbytes": 3072,
1686
+ "byteOffset": 22030336
1687
+ },
1688
+ {
1689
+ "name": "model.layers.24.post_attention_layernorm.weight",
1690
+ "shape": [
1691
+ 1536
1692
+ ],
1693
+ "dtype": "float16",
1694
+ "format": "f32-to-bf16",
1695
+ "nbytes": 3072,
1696
+ "byteOffset": 22033408
1697
+ },
1698
+ {
1699
+ "name": "model.layers.24.self_attn.c_attn.bias",
1700
+ "shape": [
1701
+ 2048
1702
+ ],
1703
+ "dtype": "float16",
1704
+ "format": "f32-to-bf16",
1705
+ "nbytes": 4096,
1706
+ "byteOffset": 22036480
1707
+ },
1708
+ {
1709
+ "name": "model.layers.24.self_attn.c_attn.weight",
1710
+ "shape": [
1711
+ 2048,
1712
+ 1536
1713
+ ],
1714
+ "dtype": "float16",
1715
+ "format": "f32-to-bf16",
1716
+ "nbytes": 6291456,
1717
+ "byteOffset": 22040576
1718
+ },
1719
+ {
1720
+ "name": "model.layers.24.self_attn.o_proj.weight",
1721
+ "shape": [
1722
+ 1536,
1723
+ 1536
1724
+ ],
1725
+ "dtype": "float16",
1726
+ "format": "f32-to-bf16",
1727
+ "nbytes": 4718592,
1728
+ "byteOffset": 28332032
1729
+ },
1730
+ {
1731
+ "name": "model.layers.25.input_layernorm.weight",
1732
+ "shape": [
1733
+ 1536
1734
+ ],
1735
+ "dtype": "float16",
1736
+ "format": "f32-to-bf16",
1737
+ "nbytes": 3072,
1738
+ "byteOffset": 33050624
1739
+ },
1740
+ {
1741
+ "name": "model.layers.25.post_attention_layernorm.weight",
1742
+ "shape": [
1743
+ 1536
1744
+ ],
1745
+ "dtype": "float16",
1746
+ "format": "f32-to-bf16",
1747
+ "nbytes": 3072,
1748
+ "byteOffset": 33053696
1749
+ },
1750
+ {
1751
+ "name": "model.layers.25.self_attn.c_attn.bias",
1752
+ "shape": [
1753
+ 2048
1754
+ ],
1755
+ "dtype": "float16",
1756
+ "format": "f32-to-bf16",
1757
+ "nbytes": 4096,
1758
+ "byteOffset": 33056768
1759
+ }
1760
+ ],
1761
+ "md5sum": "2c606ba7d834795d0792741e2744b78b"
1762
+ },
1763
+ {
1764
+ "dataPath": "params_shard_45.bin",
1765
+ "format": "raw-shard",
1766
+ "nbytes": 27525120,
1767
+ "records": [
1768
+ {
1769
+ "name": "model.layers.26.mlp.down_proj.weight",
1770
+ "shape": [
1771
+ 1536,
1772
+ 8960
1773
+ ],
1774
+ "dtype": "float16",
1775
+ "format": "f32-to-bf16",
1776
+ "nbytes": 27525120,
1777
+ "byteOffset": 0
1778
+ }
1779
+ ],
1780
+ "md5sum": "9822ca4f53df7fdfd996cf932ca71e4c"
1781
+ },
1782
+ {
1783
+ "dataPath": "params_shard_46.bin",
1784
+ "format": "raw-shard",
1785
+ "nbytes": 55050240,
1786
+ "records": [
1787
+ {
1788
+ "name": "model.layers.26.mlp.gate_up_proj.weight",
1789
+ "shape": [
1790
+ 17920,
1791
+ 1536
1792
+ ],
1793
+ "dtype": "float16",
1794
+ "format": "f32-to-bf16",
1795
+ "nbytes": 55050240,
1796
+ "byteOffset": 0
1797
+ }
1798
+ ],
1799
+ "md5sum": "ee1f23db1ec45e20fbde3a4221b65fbd"
1800
+ },
1801
+ {
1802
+ "dataPath": "params_shard_47.bin",
1803
+ "format": "raw-shard",
1804
+ "nbytes": 27525120,
1805
+ "records": [
1806
+ {
1807
+ "name": "model.layers.27.mlp.down_proj.weight",
1808
+ "shape": [
1809
+ 1536,
1810
+ 8960
1811
+ ],
1812
+ "dtype": "float16",
1813
+ "format": "f32-to-bf16",
1814
+ "nbytes": 27525120,
1815
+ "byteOffset": 0
1816
+ }
1817
+ ],
1818
+ "md5sum": "c52f61792ae29307aa852a321ff6f9f9"
1819
+ },
1820
+ {
1821
+ "dataPath": "params_shard_48.bin",
1822
+ "format": "raw-shard",
1823
+ "nbytes": 55050240,
1824
+ "records": [
1825
+ {
1826
+ "name": "model.layers.27.mlp.gate_up_proj.weight",
1827
+ "shape": [
1828
+ 17920,
1829
+ 1536
1830
+ ],
1831
+ "dtype": "float16",
1832
+ "format": "f32-to-bf16",
1833
+ "nbytes": 55050240,
1834
+ "byteOffset": 0
1835
+ }
1836
+ ],
1837
+ "md5sum": "515ae8963ca3a8122e018f2a8fc13c95"
1838
+ },
1839
+ {
1840
+ "dataPath": "params_shard_49.bin",
1841
+ "format": "raw-shard",
1842
+ "nbytes": 27525120,
1843
+ "records": [
1844
+ {
1845
+ "name": "model.layers.3.mlp.down_proj.weight",
1846
+ "shape": [
1847
+ 1536,
1848
+ 8960
1849
+ ],
1850
+ "dtype": "float16",
1851
+ "format": "f32-to-bf16",
1852
+ "nbytes": 27525120,
1853
+ "byteOffset": 0
1854
+ }
1855
+ ],
1856
+ "md5sum": "4885a9b7a3a5939478bfe31cee3a5b28"
1857
+ },
1858
+ {
1859
+ "dataPath": "params_shard_50.bin",
1860
+ "format": "raw-shard",
1861
+ "nbytes": 55050240,
1862
+ "records": [
1863
+ {
1864
+ "name": "model.layers.3.mlp.gate_up_proj.weight",
1865
+ "shape": [
1866
+ 17920,
1867
+ 1536
1868
+ ],
1869
+ "dtype": "float16",
1870
+ "format": "f32-to-bf16",
1871
+ "nbytes": 55050240,
1872
+ "byteOffset": 0
1873
+ }
1874
+ ],
1875
+ "md5sum": "7a09a1539e93c381c1a48fb995b6bbfd"
1876
+ },
1877
+ {
1878
+ "dataPath": "params_shard_51.bin",
1879
+ "format": "raw-shard",
1880
+ "nbytes": 33060864,
1881
+ "records": [
1882
+ {
1883
+ "name": "model.layers.25.self_attn.c_attn.weight",
1884
+ "shape": [
1885
+ 2048,
1886
+ 1536
1887
+ ],
1888
+ "dtype": "float16",
1889
+ "format": "f32-to-bf16",
1890
+ "nbytes": 6291456,
1891
+ "byteOffset": 0
1892
+ },
1893
+ {
1894
+ "name": "model.layers.25.self_attn.o_proj.weight",
1895
+ "shape": [
1896
+ 1536,
1897
+ 1536
1898
+ ],
1899
+ "dtype": "float16",
1900
+ "format": "f32-to-bf16",
1901
+ "nbytes": 4718592,
1902
+ "byteOffset": 6291456
1903
+ },
1904
+ {
1905
+ "name": "model.layers.26.input_layernorm.weight",
1906
+ "shape": [
1907
+ 1536
1908
+ ],
1909
+ "dtype": "float16",
1910
+ "format": "f32-to-bf16",
1911
+ "nbytes": 3072,
1912
+ "byteOffset": 11010048
1913
+ },
1914
+ {
1915
+ "name": "model.layers.26.post_attention_layernorm.weight",
1916
+ "shape": [
1917
+ 1536
1918
+ ],
1919
+ "dtype": "float16",
1920
+ "format": "f32-to-bf16",
1921
+ "nbytes": 3072,
1922
+ "byteOffset": 11013120
1923
+ },
1924
+ {
1925
+ "name": "model.layers.26.self_attn.c_attn.bias",
1926
+ "shape": [
1927
+ 2048
1928
+ ],
1929
+ "dtype": "float16",
1930
+ "format": "f32-to-bf16",
1931
+ "nbytes": 4096,
1932
+ "byteOffset": 11016192
1933
+ },
1934
+ {
1935
+ "name": "model.layers.26.self_attn.c_attn.weight",
1936
+ "shape": [
1937
+ 2048,
1938
+ 1536
1939
+ ],
1940
+ "dtype": "float16",
1941
+ "format": "f32-to-bf16",
1942
+ "nbytes": 6291456,
1943
+ "byteOffset": 11020288
1944
+ },
1945
+ {
1946
+ "name": "model.layers.26.self_attn.o_proj.weight",
1947
+ "shape": [
1948
+ 1536,
1949
+ 1536
1950
+ ],
1951
+ "dtype": "float16",
1952
+ "format": "f32-to-bf16",
1953
+ "nbytes": 4718592,
1954
+ "byteOffset": 17311744
1955
+ },
1956
+ {
1957
+ "name": "model.layers.27.input_layernorm.weight",
1958
+ "shape": [
1959
+ 1536
1960
+ ],
1961
+ "dtype": "float16",
1962
+ "format": "f32-to-bf16",
1963
+ "nbytes": 3072,
1964
+ "byteOffset": 22030336
1965
+ },
1966
+ {
1967
+ "name": "model.layers.27.post_attention_layernorm.weight",
1968
+ "shape": [
1969
+ 1536
1970
+ ],
1971
+ "dtype": "float16",
1972
+ "format": "f32-to-bf16",
1973
+ "nbytes": 3072,
1974
+ "byteOffset": 22033408
1975
+ },
1976
+ {
1977
+ "name": "model.layers.27.self_attn.c_attn.bias",
1978
+ "shape": [
1979
+ 2048
1980
+ ],
1981
+ "dtype": "float16",
1982
+ "format": "f32-to-bf16",
1983
+ "nbytes": 4096,
1984
+ "byteOffset": 22036480
1985
+ },
1986
+ {
1987
+ "name": "model.layers.27.self_attn.c_attn.weight",
1988
+ "shape": [
1989
+ 2048,
1990
+ 1536
1991
+ ],
1992
+ "dtype": "float16",
1993
+ "format": "f32-to-bf16",
1994
+ "nbytes": 6291456,
1995
+ "byteOffset": 22040576
1996
+ },
1997
+ {
1998
+ "name": "model.layers.27.self_attn.o_proj.weight",
1999
+ "shape": [
2000
+ 1536,
2001
+ 1536
2002
+ ],
2003
+ "dtype": "float16",
2004
+ "format": "f32-to-bf16",
2005
+ "nbytes": 4718592,
2006
+ "byteOffset": 28332032
2007
+ },
2008
+ {
2009
+ "name": "model.layers.3.input_layernorm.weight",
2010
+ "shape": [
2011
+ 1536
2012
+ ],
2013
+ "dtype": "float16",
2014
+ "format": "f32-to-bf16",
2015
+ "nbytes": 3072,
2016
+ "byteOffset": 33050624
2017
+ },
2018
+ {
2019
+ "name": "model.layers.3.post_attention_layernorm.weight",
2020
+ "shape": [
2021
+ 1536
2022
+ ],
2023
+ "dtype": "float16",
2024
+ "format": "f32-to-bf16",
2025
+ "nbytes": 3072,
2026
+ "byteOffset": 33053696
2027
+ },
2028
+ {
2029
+ "name": "model.layers.3.self_attn.c_attn.bias",
2030
+ "shape": [
2031
+ 2048
2032
+ ],
2033
+ "dtype": "float16",
2034
+ "format": "f32-to-bf16",
2035
+ "nbytes": 4096,
2036
+ "byteOffset": 33056768
2037
+ }
2038
+ ],
2039
+ "md5sum": "42f20e01ea60ca0c560a6beac3595d37"
2040
+ },
2041
+ {
2042
+ "dataPath": "params_shard_52.bin",
2043
+ "format": "raw-shard",
2044
+ "nbytes": 27525120,
2045
+ "records": [
2046
+ {
2047
+ "name": "model.layers.4.mlp.down_proj.weight",
2048
+ "shape": [
2049
+ 1536,
2050
+ 8960
2051
+ ],
2052
+ "dtype": "float16",
2053
+ "format": "f32-to-bf16",
2054
+ "nbytes": 27525120,
2055
+ "byteOffset": 0
2056
+ }
2057
+ ],
2058
+ "md5sum": "a3e225fc984284ae73ce70439ea4909d"
2059
+ },
2060
+ {
2061
+ "dataPath": "params_shard_53.bin",
2062
+ "format": "raw-shard",
2063
+ "nbytes": 55050240,
2064
+ "records": [
2065
+ {
2066
+ "name": "model.layers.4.mlp.gate_up_proj.weight",
2067
+ "shape": [
2068
+ 17920,
2069
+ 1536
2070
+ ],
2071
+ "dtype": "float16",
2072
+ "format": "f32-to-bf16",
2073
+ "nbytes": 55050240,
2074
+ "byteOffset": 0
2075
+ }
2076
+ ],
2077
+ "md5sum": "4d86a4ac8c5b1e0f6f38624a5ee0e7bf"
2078
+ },
2079
+ {
2080
+ "dataPath": "params_shard_54.bin",
2081
+ "format": "raw-shard",
2082
+ "nbytes": 27525120,
2083
+ "records": [
2084
+ {
2085
+ "name": "model.layers.5.mlp.down_proj.weight",
2086
+ "shape": [
2087
+ 1536,
2088
+ 8960
2089
+ ],
2090
+ "dtype": "float16",
2091
+ "format": "f32-to-bf16",
2092
+ "nbytes": 27525120,
2093
+ "byteOffset": 0
2094
+ }
2095
+ ],
2096
+ "md5sum": "5372e6b4fb297a19e1c9862127797e94"
2097
+ },
2098
+ {
2099
+ "dataPath": "params_shard_55.bin",
2100
+ "format": "raw-shard",
2101
+ "nbytes": 55050240,
2102
+ "records": [
2103
+ {
2104
+ "name": "model.layers.5.mlp.gate_up_proj.weight",
2105
+ "shape": [
2106
+ 17920,
2107
+ 1536
2108
+ ],
2109
+ "dtype": "float16",
2110
+ "format": "f32-to-bf16",
2111
+ "nbytes": 55050240,
2112
+ "byteOffset": 0
2113
+ }
2114
+ ],
2115
+ "md5sum": "41b9625467615c755e3fdcd20ec4e53c"
2116
+ },
2117
+ {
2118
+ "dataPath": "params_shard_56.bin",
2119
+ "format": "raw-shard",
2120
+ "nbytes": 27525120,
2121
+ "records": [
2122
+ {
2123
+ "name": "model.layers.6.mlp.down_proj.weight",
2124
+ "shape": [
2125
+ 1536,
2126
+ 8960
2127
+ ],
2128
+ "dtype": "float16",
2129
+ "format": "f32-to-bf16",
2130
+ "nbytes": 27525120,
2131
+ "byteOffset": 0
2132
+ }
2133
+ ],
2134
+ "md5sum": "2f7cd67584bbecd1c184e87d0ee6ba06"
2135
+ },
2136
+ {
2137
+ "dataPath": "params_shard_57.bin",
2138
+ "format": "raw-shard",
2139
+ "nbytes": 55050240,
2140
+ "records": [
2141
+ {
2142
+ "name": "model.layers.6.mlp.gate_up_proj.weight",
2143
+ "shape": [
2144
+ 17920,
2145
+ 1536
2146
+ ],
2147
+ "dtype": "float16",
2148
+ "format": "f32-to-bf16",
2149
+ "nbytes": 55050240,
2150
+ "byteOffset": 0
2151
+ }
2152
+ ],
2153
+ "md5sum": "69bf0fb11d59922a5369f782851407b2"
2154
+ },
2155
+ {
2156
+ "dataPath": "params_shard_58.bin",
2157
+ "format": "raw-shard",
2158
+ "nbytes": 33060864,
2159
+ "records": [
2160
+ {
2161
+ "name": "model.layers.3.self_attn.c_attn.weight",
2162
+ "shape": [
2163
+ 2048,
2164
+ 1536
2165
+ ],
2166
+ "dtype": "float16",
2167
+ "format": "f32-to-bf16",
2168
+ "nbytes": 6291456,
2169
+ "byteOffset": 0
2170
+ },
2171
+ {
2172
+ "name": "model.layers.3.self_attn.o_proj.weight",
2173
+ "shape": [
2174
+ 1536,
2175
+ 1536
2176
+ ],
2177
+ "dtype": "float16",
2178
+ "format": "f32-to-bf16",
2179
+ "nbytes": 4718592,
2180
+ "byteOffset": 6291456
2181
+ },
2182
+ {
2183
+ "name": "model.layers.4.input_layernorm.weight",
2184
+ "shape": [
2185
+ 1536
2186
+ ],
2187
+ "dtype": "float16",
2188
+ "format": "f32-to-bf16",
2189
+ "nbytes": 3072,
2190
+ "byteOffset": 11010048
2191
+ },
2192
+ {
2193
+ "name": "model.layers.4.post_attention_layernorm.weight",
2194
+ "shape": [
2195
+ 1536
2196
+ ],
2197
+ "dtype": "float16",
2198
+ "format": "f32-to-bf16",
2199
+ "nbytes": 3072,
2200
+ "byteOffset": 11013120
2201
+ },
2202
+ {
2203
+ "name": "model.layers.4.self_attn.c_attn.bias",
2204
+ "shape": [
2205
+ 2048
2206
+ ],
2207
+ "dtype": "float16",
2208
+ "format": "f32-to-bf16",
2209
+ "nbytes": 4096,
2210
+ "byteOffset": 11016192
2211
+ },
2212
+ {
2213
+ "name": "model.layers.4.self_attn.c_attn.weight",
2214
+ "shape": [
2215
+ 2048,
2216
+ 1536
2217
+ ],
2218
+ "dtype": "float16",
2219
+ "format": "f32-to-bf16",
2220
+ "nbytes": 6291456,
2221
+ "byteOffset": 11020288
2222
+ },
2223
+ {
2224
+ "name": "model.layers.4.self_attn.o_proj.weight",
2225
+ "shape": [
2226
+ 1536,
2227
+ 1536
2228
+ ],
2229
+ "dtype": "float16",
2230
+ "format": "f32-to-bf16",
2231
+ "nbytes": 4718592,
2232
+ "byteOffset": 17311744
2233
+ },
2234
+ {
2235
+ "name": "model.layers.5.input_layernorm.weight",
2236
+ "shape": [
2237
+ 1536
2238
+ ],
2239
+ "dtype": "float16",
2240
+ "format": "f32-to-bf16",
2241
+ "nbytes": 3072,
2242
+ "byteOffset": 22030336
2243
+ },
2244
+ {
2245
+ "name": "model.layers.5.post_attention_layernorm.weight",
2246
+ "shape": [
2247
+ 1536
2248
+ ],
2249
+ "dtype": "float16",
2250
+ "format": "f32-to-bf16",
2251
+ "nbytes": 3072,
2252
+ "byteOffset": 22033408
2253
+ },
2254
+ {
2255
+ "name": "model.layers.5.self_attn.c_attn.bias",
2256
+ "shape": [
2257
+ 2048
2258
+ ],
2259
+ "dtype": "float16",
2260
+ "format": "f32-to-bf16",
2261
+ "nbytes": 4096,
2262
+ "byteOffset": 22036480
2263
+ },
2264
+ {
2265
+ "name": "model.layers.5.self_attn.c_attn.weight",
2266
+ "shape": [
2267
+ 2048,
2268
+ 1536
2269
+ ],
2270
+ "dtype": "float16",
2271
+ "format": "f32-to-bf16",
2272
+ "nbytes": 6291456,
2273
+ "byteOffset": 22040576
2274
+ },
2275
+ {
2276
+ "name": "model.layers.5.self_attn.o_proj.weight",
2277
+ "shape": [
2278
+ 1536,
2279
+ 1536
2280
+ ],
2281
+ "dtype": "float16",
2282
+ "format": "f32-to-bf16",
2283
+ "nbytes": 4718592,
2284
+ "byteOffset": 28332032
2285
+ },
2286
+ {
2287
+ "name": "model.layers.6.input_layernorm.weight",
2288
+ "shape": [
2289
+ 1536
2290
+ ],
2291
+ "dtype": "float16",
2292
+ "format": "f32-to-bf16",
2293
+ "nbytes": 3072,
2294
+ "byteOffset": 33050624
2295
+ },
2296
+ {
2297
+ "name": "model.layers.6.post_attention_layernorm.weight",
2298
+ "shape": [
2299
+ 1536
2300
+ ],
2301
+ "dtype": "float16",
2302
+ "format": "f32-to-bf16",
2303
+ "nbytes": 3072,
2304
+ "byteOffset": 33053696
2305
+ },
2306
+ {
2307
+ "name": "model.layers.6.self_attn.c_attn.bias",
2308
+ "shape": [
2309
+ 2048
2310
+ ],
2311
+ "dtype": "float16",
2312
+ "format": "f32-to-bf16",
2313
+ "nbytes": 4096,
2314
+ "byteOffset": 33056768
2315
+ }
2316
+ ],
2317
+ "md5sum": "47b60369b87e10f488935fb5f4c33ddb"
2318
+ },
2319
+ {
2320
+ "dataPath": "params_shard_59.bin",
2321
+ "format": "raw-shard",
2322
+ "nbytes": 27525120,
2323
+ "records": [
2324
+ {
2325
+ "name": "model.layers.7.mlp.down_proj.weight",
2326
+ "shape": [
2327
+ 1536,
2328
+ 8960
2329
+ ],
2330
+ "dtype": "float16",
2331
+ "format": "f32-to-bf16",
2332
+ "nbytes": 27525120,
2333
+ "byteOffset": 0
2334
+ }
2335
+ ],
2336
+ "md5sum": "6f92932c3dafdfd2d6b579f22644d950"
2337
+ },
2338
+ {
2339
+ "dataPath": "params_shard_60.bin",
2340
+ "format": "raw-shard",
2341
+ "nbytes": 55050240,
2342
+ "records": [
2343
+ {
2344
+ "name": "model.layers.7.mlp.gate_up_proj.weight",
2345
+ "shape": [
2346
+ 17920,
2347
+ 1536
2348
+ ],
2349
+ "dtype": "float16",
2350
+ "format": "f32-to-bf16",
2351
+ "nbytes": 55050240,
2352
+ "byteOffset": 0
2353
+ }
2354
+ ],
2355
+ "md5sum": "6ccdaa66cc9db71bf6cb0a14457fd073"
2356
+ },
2357
+ {
2358
+ "dataPath": "params_shard_61.bin",
2359
+ "format": "raw-shard",
2360
+ "nbytes": 27525120,
2361
+ "records": [
2362
+ {
2363
+ "name": "model.layers.8.mlp.down_proj.weight",
2364
+ "shape": [
2365
+ 1536,
2366
+ 8960
2367
+ ],
2368
+ "dtype": "float16",
2369
+ "format": "f32-to-bf16",
2370
+ "nbytes": 27525120,
2371
+ "byteOffset": 0
2372
+ }
2373
+ ],
2374
+ "md5sum": "f5745c3c21c80053544dce266c7114dd"
2375
+ },
2376
+ {
2377
+ "dataPath": "params_shard_62.bin",
2378
+ "format": "raw-shard",
2379
+ "nbytes": 55050240,
2380
+ "records": [
2381
+ {
2382
+ "name": "model.layers.8.mlp.gate_up_proj.weight",
2383
+ "shape": [
2384
+ 17920,
2385
+ 1536
2386
+ ],
2387
+ "dtype": "float16",
2388
+ "format": "f32-to-bf16",
2389
+ "nbytes": 55050240,
2390
+ "byteOffset": 0
2391
+ }
2392
+ ],
2393
+ "md5sum": "22bb273ed7496d6748489b7fe0b503fa"
2394
+ },
2395
+ {
2396
+ "dataPath": "params_shard_63.bin",
2397
+ "format": "raw-shard",
2398
+ "nbytes": 27525120,
2399
+ "records": [
2400
+ {
2401
+ "name": "model.layers.9.mlp.down_proj.weight",
2402
+ "shape": [
2403
+ 1536,
2404
+ 8960
2405
+ ],
2406
+ "dtype": "float16",
2407
+ "format": "f32-to-bf16",
2408
+ "nbytes": 27525120,
2409
+ "byteOffset": 0
2410
+ }
2411
+ ],
2412
+ "md5sum": "21f7faf648e423db9aa908d3a558ef2a"
2413
+ },
2414
+ {
2415
+ "dataPath": "params_shard_64.bin",
2416
+ "format": "raw-shard",
2417
+ "nbytes": 55050240,
2418
+ "records": [
2419
+ {
2420
+ "name": "model.layers.9.mlp.gate_up_proj.weight",
2421
+ "shape": [
2422
+ 17920,
2423
+ 1536
2424
+ ],
2425
+ "dtype": "float16",
2426
+ "format": "f32-to-bf16",
2427
+ "nbytes": 55050240,
2428
+ "byteOffset": 0
2429
+ }
2430
+ ],
2431
+ "md5sum": "778c6c84bac2ed0e9c194a06c0c61bdd"
2432
+ },
2433
+ {
2434
+ "dataPath": "params_shard_65.bin",
2435
+ "format": "raw-shard",
2436
+ "nbytes": 33060864,
2437
+ "records": [
2438
+ {
2439
+ "name": "model.layers.6.self_attn.c_attn.weight",
2440
+ "shape": [
2441
+ 2048,
2442
+ 1536
2443
+ ],
2444
+ "dtype": "float16",
2445
+ "format": "f32-to-bf16",
2446
+ "nbytes": 6291456,
2447
+ "byteOffset": 0
2448
+ },
2449
+ {
2450
+ "name": "model.layers.6.self_attn.o_proj.weight",
2451
+ "shape": [
2452
+ 1536,
2453
+ 1536
2454
+ ],
2455
+ "dtype": "float16",
2456
+ "format": "f32-to-bf16",
2457
+ "nbytes": 4718592,
2458
+ "byteOffset": 6291456
2459
+ },
2460
+ {
2461
+ "name": "model.layers.7.input_layernorm.weight",
2462
+ "shape": [
2463
+ 1536
2464
+ ],
2465
+ "dtype": "float16",
2466
+ "format": "f32-to-bf16",
2467
+ "nbytes": 3072,
2468
+ "byteOffset": 11010048
2469
+ },
2470
+ {
2471
+ "name": "model.layers.7.post_attention_layernorm.weight",
2472
+ "shape": [
2473
+ 1536
2474
+ ],
2475
+ "dtype": "float16",
2476
+ "format": "f32-to-bf16",
2477
+ "nbytes": 3072,
2478
+ "byteOffset": 11013120
2479
+ },
2480
+ {
2481
+ "name": "model.layers.7.self_attn.c_attn.bias",
2482
+ "shape": [
2483
+ 2048
2484
+ ],
2485
+ "dtype": "float16",
2486
+ "format": "f32-to-bf16",
2487
+ "nbytes": 4096,
2488
+ "byteOffset": 11016192
2489
+ },
2490
+ {
2491
+ "name": "model.layers.7.self_attn.c_attn.weight",
2492
+ "shape": [
2493
+ 2048,
2494
+ 1536
2495
+ ],
2496
+ "dtype": "float16",
2497
+ "format": "f32-to-bf16",
2498
+ "nbytes": 6291456,
2499
+ "byteOffset": 11020288
2500
+ },
2501
+ {
2502
+ "name": "model.layers.7.self_attn.o_proj.weight",
2503
+ "shape": [
2504
+ 1536,
2505
+ 1536
2506
+ ],
2507
+ "dtype": "float16",
2508
+ "format": "f32-to-bf16",
2509
+ "nbytes": 4718592,
2510
+ "byteOffset": 17311744
2511
+ },
2512
+ {
2513
+ "name": "model.layers.8.input_layernorm.weight",
2514
+ "shape": [
2515
+ 1536
2516
+ ],
2517
+ "dtype": "float16",
2518
+ "format": "f32-to-bf16",
2519
+ "nbytes": 3072,
2520
+ "byteOffset": 22030336
2521
+ },
2522
+ {
2523
+ "name": "model.layers.8.post_attention_layernorm.weight",
2524
+ "shape": [
2525
+ 1536
2526
+ ],
2527
+ "dtype": "float16",
2528
+ "format": "f32-to-bf16",
2529
+ "nbytes": 3072,
2530
+ "byteOffset": 22033408
2531
+ },
2532
+ {
2533
+ "name": "model.layers.8.self_attn.c_attn.bias",
2534
+ "shape": [
2535
+ 2048
2536
+ ],
2537
+ "dtype": "float16",
2538
+ "format": "f32-to-bf16",
2539
+ "nbytes": 4096,
2540
+ "byteOffset": 22036480
2541
+ },
2542
+ {
2543
+ "name": "model.layers.8.self_attn.c_attn.weight",
2544
+ "shape": [
2545
+ 2048,
2546
+ 1536
2547
+ ],
2548
+ "dtype": "float16",
2549
+ "format": "f32-to-bf16",
2550
+ "nbytes": 6291456,
2551
+ "byteOffset": 22040576
2552
+ },
2553
+ {
2554
+ "name": "model.layers.8.self_attn.o_proj.weight",
2555
+ "shape": [
2556
+ 1536,
2557
+ 1536
2558
+ ],
2559
+ "dtype": "float16",
2560
+ "format": "f32-to-bf16",
2561
+ "nbytes": 4718592,
2562
+ "byteOffset": 28332032
2563
+ },
2564
+ {
2565
+ "name": "model.layers.9.input_layernorm.weight",
2566
+ "shape": [
2567
+ 1536
2568
+ ],
2569
+ "dtype": "float16",
2570
+ "format": "f32-to-bf16",
2571
+ "nbytes": 3072,
2572
+ "byteOffset": 33050624
2573
+ },
2574
+ {
2575
+ "name": "model.layers.9.post_attention_layernorm.weight",
2576
+ "shape": [
2577
+ 1536
2578
+ ],
2579
+ "dtype": "float16",
2580
+ "format": "f32-to-bf16",
2581
+ "nbytes": 3072,
2582
+ "byteOffset": 33053696
2583
+ },
2584
+ {
2585
+ "name": "model.layers.9.self_attn.c_attn.bias",
2586
+ "shape": [
2587
+ 2048
2588
+ ],
2589
+ "dtype": "float16",
2590
+ "format": "f32-to-bf16",
2591
+ "nbytes": 4096,
2592
+ "byteOffset": 33056768
2593
+ }
2594
+ ],
2595
+ "md5sum": "ab7b46752eff1100c8eed4cd1269dfac"
2596
+ },
2597
+ {
2598
+ "dataPath": "params_shard_66.bin",
2599
+ "format": "raw-shard",
2600
+ "nbytes": 11013120,
2601
+ "records": [
2602
+ {
2603
+ "name": "model.layers.9.self_attn.c_attn.weight",
2604
+ "shape": [
2605
+ 2048,
2606
+ 1536
2607
+ ],
2608
+ "dtype": "float16",
2609
+ "format": "f32-to-bf16",
2610
+ "nbytes": 6291456,
2611
+ "byteOffset": 0
2612
+ },
2613
+ {
2614
+ "name": "model.layers.9.self_attn.o_proj.weight",
2615
+ "shape": [
2616
+ 1536,
2617
+ 1536
2618
+ ],
2619
+ "dtype": "float16",
2620
+ "format": "f32-to-bf16",
2621
+ "nbytes": 4718592,
2622
+ "byteOffset": 6291456
2623
+ },
2624
+ {
2625
+ "name": "model.norm.weight",
2626
+ "shape": [
2627
+ 1536
2628
+ ],
2629
+ "dtype": "float16",
2630
+ "format": "f32-to-bf16",
2631
+ "nbytes": 3072,
2632
+ "byteOffset": 11010048
2633
+ }
2634
+ ],
2635
+ "md5sum": "563b659b7554e11e85fbb315b50edce0"
2636
+ }
2637
+ ]
2638
+ }
params_shard_0.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e52e67f3e01b3b2a2e781a630b8c210d65068019185bf2962837a0b963e3508d
3
+ size 466747392
params_shard_1.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3b68b8c43006b1d90798f1e593f711a37bdf827c7489c90e221f7bc0a59662f8
3
+ size 55050240
params_shard_10.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef77f2f2f8371ea55d4826af1791b74b630f29e3d59a46a1895daa2096690f3a
3
+ size 27525120
params_shard_11.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4bed86af132dd21fd8bcee5ee19b231432b8f4308a3a18a259a7ce6a41387355
3
+ size 55050240
params_shard_12.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a44d23f481b99bf7a751ea93de16db9ee241594a9110b926640ad55701883211
3
+ size 27525120
params_shard_13.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e6b4d45ca1ae2a26fdc0e248825eba5a72af7a64c5b05c3eff3899d2e3a1606d
3
+ size 55050240
params_shard_14.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:596d8932becd3531e0e966845a9aa7e834b9460f778078efb238dd5e195aafe6
3
+ size 27525120
params_shard_15.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:374c6996ee87b52545005f3bd37e383794e456d645b01b3d5b548b12420bb7bc
3
+ size 55050240
params_shard_16.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4ca7a2788f6a210dd481c0fdf02d8c25fca13ee9394e86a17164cb259ca9b686
3
+ size 33060864
params_shard_17.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:242617ec60c4164ca255b285223cb5118c84763283d632bf78504233f03ef840
3
+ size 27525120
params_shard_18.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ccfcf1637ad461bd2d03e6e9571bf7dce160b4540cc2da9d2403a7efc282afcf
3
+ size 55050240
params_shard_19.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e20409ea1c3ad026039b5895e8e2b4b6aacab2ff456f50150b772bef0dd9ddc2
3
+ size 27525120
params_shard_2.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c2038754e4eeb80507875ce5b263db60e67c84c921128159d125c312a8c95955
3
+ size 27535360
params_shard_20.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c95b91799bf1753eb8d35e4751f1b5355bf26ee8c2d657fc4b01771853d0f23f
3
+ size 55050240
params_shard_21.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ad41206d73eed0c1c40498cf4d334a6ac4de7b1f31896a69f4f2e66375e299ae
3
+ size 27525120
params_shard_22.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:03c8093fba7b514de5e84e4f87b0d97d90e6b30478468ff6d2c99b87197d9cf6
3
+ size 55050240
params_shard_23.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6519693e01350c94faa021bf53968dbfd6782311047f7daf16f4a0e4be81bb3
3
+ size 33060864
params_shard_24.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c08ee0fa38fc72bebd367324caa461a8a0e446c4fbf4c657b59c758fa574d16b
3
+ size 27525120
params_shard_25.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a25b17884c1487063deec78d90fa38f3d3eb509fa664841ed55cc958ffee7e8c
3
+ size 55050240
params_shard_26.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ce991028e815f39ec9a848d4e1d35678acafe55dd87d36daf781096ffaa49c8
3
+ size 27525120
params_shard_27.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef47567c11153c814d1354d608ba1f3b2836b003678d7ded7d609d38acd39f48
3
+ size 55050240
params_shard_28.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1be370f53070024a6d7ac3fd25baeb40fc0cb1ea04398c9f57fbe63b104d7ceb
3
+ size 27525120
params_shard_29.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b10574aeb6390454248ce007fab8457bdb9a057a5ff45cafe099bd8233cf2ae5
3
+ size 55050240
params_shard_3.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9df561eded03a577e690407b455a96c5f3c553b84ee52f5c2a27910b6f178816
3
+ size 27525120
params_shard_30.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e699927ff2b1e533e6a41a939bcf78a23ee0f58d78611259707fb09ae0b54c4d
3
+ size 33060864
params_shard_31.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6fc578b457b8c932692e542246f8561f86dbfc67b2b774b4ffe99856402bb752
3
+ size 27525120
params_shard_32.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:927ad93100f5d85acd70a8b66c1375ecd19e0b160939ef5387d43801b32b6341
3
+ size 55050240
params_shard_33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9a0f591e3c8cc2240d5992d39eb662d10cbb8b02ddffaa7cdb7b926cc7ce7120
3
+ size 27525120
params_shard_34.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70a0172d98e107fdf59939016a4b29859c88de3b4443f4bd727fde02525e0936
3
+ size 55050240
params_shard_35.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e5b984e9c2d9873e14b92525667d7e7f3e071fc1da61e768abc45231c61e8f4
3
+ size 27525120
params_shard_36.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:237da286579261cc345ff1ef841f642e6bbb0c84fd586989bf9eb58f577526e3
3
+ size 55050240
params_shard_37.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4af1bc6340aed5ce8d9bd207373bee0d46547ee1f9fe75ef12b88b98f01e30a9
3
+ size 33060864
params_shard_38.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:09552a4e2aafd4dcdfb50e688124a51f760b0aa69421f1b46b1566bc260b444a
3
+ size 27525120
params_shard_39.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:287d888182bbb33aec90f1f7b00d771e7a4299d9b4dc2c38af836b7c6c78eb75
3
+ size 55050240
params_shard_4.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d497dc56086c8994dfff7e8f2d0787c01267028d5ec064b11119526da83682b8
3
+ size 55050240
params_shard_40.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5dadbbde72435bb9d77a8719c67743e082d3aa729bc6ae5866a1c22347a9c067
3
+ size 27525120
params_shard_41.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89e6b0bd9bf8ffad82d6ae3b915084ca96e12fdc24bb95ec14eba6211597806d
3
+ size 55050240
params_shard_42.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e2625fc15b6edb98b78908502409184f2926165646d58f19ee47eaa0d8a8b9e
3
+ size 27525120
params_shard_43.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:44d695cc5b6592e7acb83edeafc5e0b86e0ce8d6050cf871d22cf76828b51443
3
+ size 55050240
params_shard_44.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed7d66e1ed271919c67859d557f283f5f8098bc877445e1786efb675a4334ad1
3
+ size 33060864
params_shard_45.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:63703c93832904fb8d350f0df9c17f843db9fdc12280c427da712974aec14621
3
+ size 27525120
params_shard_46.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:02088c12a314b274c23bf214d4e522efa67719b28b08cb383393922b7657117f
3
+ size 55050240
params_shard_47.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ad5ec80ae54f798b0067bf5859f06cc130bef7b8cfcffff143a766234c085a7
3
+ size 27525120
params_shard_48.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e6aba263bcb61253c73e2d0426d9498a3df78480b9386268ec0b207f6db5410
3
+ size 55050240
params_shard_49.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5db16f591a041cafd5ba338850ae3fac605b618227d87d6463487f3857858bde
3
+ size 27525120
params_shard_5.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b246b032600815a1eacc48eb5367e931c3075f645b14853d7a17f54f915de158
3
+ size 27525120
params_shard_50.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8f82c57c066e841ab7f6dc47fd60536661dee0ccb30b5716ee755c0743baf4f8
3
+ size 55050240