riczhou commited on
Commit
bf57713
·
verified ·
1 Parent(s): 3fe05a6

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
mlc-chat-config.json ADDED
@@ -0,0 +1,76 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": 10000,
14
+ "vocab_size": 151936,
15
+ "tie_word_embeddings": false,
16
+ "context_window_size": 131072,
17
+ "prefill_chunk_size": 8192,
18
+ "tensor_parallel_shards": 1,
19
+ "head_dim": 128,
20
+ "dtype": "float32",
21
+ "max_batch_size": 128
22
+ },
23
+ "vocab_size": 151936,
24
+ "context_window_size": 131072,
25
+ "sliding_window_size": -1,
26
+ "prefill_chunk_size": 8192,
27
+ "attention_sink_size": -1,
28
+ "tensor_parallel_shards": 1,
29
+ "pipeline_parallel_stages": 1,
30
+ "temperature": 1.0,
31
+ "presence_penalty": 0.0,
32
+ "frequency_penalty": 0.0,
33
+ "repetition_penalty": 1.0,
34
+ "top_p": 1.0,
35
+ "tokenizer_files": [
36
+ "tokenizer.json",
37
+ "tokenizer_config.json"
38
+ ],
39
+ "tokenizer_info": {
40
+ "token_postproc_method": "byte_level",
41
+ "prepend_space_in_encode": false,
42
+ "strip_space_in_decode": false
43
+ },
44
+ "conv_template": {
45
+ "name": "deepseek_v3",
46
+ "system_template": "<\uff5cbegin\u2581of\u2581sentence\uff5c>{system_message}",
47
+ "system_message": "You are a helpful assistant.",
48
+ "system_prefix_token_ids": null,
49
+ "add_role_after_system_message": true,
50
+ "roles": {
51
+ "user": "<\uff5cUser\uff5c>",
52
+ "assistant": "<\uff5cAssistant\uff5c>"
53
+ },
54
+ "role_templates": {
55
+ "user": "{user_message}",
56
+ "assistant": "{assistant_message}",
57
+ "tool": "{tool_message}"
58
+ },
59
+ "messages": [],
60
+ "seps": [
61
+ "",
62
+ "<\uff5cend\u2581of\u2581sentence\uff5c>"
63
+ ],
64
+ "role_content_sep": "",
65
+ "role_empty_sep": "",
66
+ "stop_str": [],
67
+ "stop_token_ids": [
68
+ 151643
69
+ ],
70
+ "function_string": "",
71
+ "use_function_calling": false
72
+ },
73
+ "pad_token_id": 0,
74
+ "bos_token_id": 151643,
75
+ "eos_token_id": 151643
76
+ }
ndarray-cache.json ADDED
@@ -0,0 +1,2657 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "ParamSize": 199,
4
+ "ParamBytes": 3554176000.0,
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": "lm_head.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": "4d8a904b17c498e9785aa867e17d7c62"
26
+ },
27
+ {
28
+ "dataPath": "params_shard_1.bin",
29
+ "format": "raw-shard",
30
+ "nbytes": 466747392,
31
+ "records": [
32
+ {
33
+ "name": "model.embed_tokens.weight",
34
+ "shape": [
35
+ 151936,
36
+ 1536
37
+ ],
38
+ "dtype": "float16",
39
+ "format": "f32-to-bf16",
40
+ "nbytes": 466747392,
41
+ "byteOffset": 0
42
+ }
43
+ ],
44
+ "md5sum": "22e897f1f35081a84326de4fde176669"
45
+ },
46
+ {
47
+ "dataPath": "params_shard_2.bin",
48
+ "format": "raw-shard",
49
+ "nbytes": 55050240,
50
+ "records": [
51
+ {
52
+ "name": "model.layers.0.mlp.gate_up_proj.weight",
53
+ "shape": [
54
+ 17920,
55
+ 1536
56
+ ],
57
+ "dtype": "float16",
58
+ "format": "f32-to-bf16",
59
+ "nbytes": 55050240,
60
+ "byteOffset": 0
61
+ }
62
+ ],
63
+ "md5sum": "81995238ec09105c42af2ff9cf79fc81"
64
+ },
65
+ {
66
+ "dataPath": "params_shard_3.bin",
67
+ "format": "raw-shard",
68
+ "nbytes": 27535360,
69
+ "records": [
70
+ {
71
+ "name": "model.layers.0.input_layernorm.weight",
72
+ "shape": [
73
+ 1536
74
+ ],
75
+ "dtype": "float16",
76
+ "format": "f32-to-bf16",
77
+ "nbytes": 3072,
78
+ "byteOffset": 0
79
+ },
80
+ {
81
+ "name": "model.layers.0.mlp.down_proj.weight",
82
+ "shape": [
83
+ 1536,
84
+ 8960
85
+ ],
86
+ "dtype": "float16",
87
+ "format": "f32-to-bf16",
88
+ "nbytes": 27525120,
89
+ "byteOffset": 3072
90
+ },
91
+ {
92
+ "name": "model.layers.0.post_attention_layernorm.weight",
93
+ "shape": [
94
+ 1536
95
+ ],
96
+ "dtype": "float16",
97
+ "format": "f32-to-bf16",
98
+ "nbytes": 3072,
99
+ "byteOffset": 27528192
100
+ },
101
+ {
102
+ "name": "model.layers.0.self_attn.c_attn.bias",
103
+ "shape": [
104
+ 2048
105
+ ],
106
+ "dtype": "float16",
107
+ "format": "f32-to-bf16",
108
+ "nbytes": 4096,
109
+ "byteOffset": 27531264
110
+ }
111
+ ],
112
+ "md5sum": "1a2e1e13b84dfdee8d5957a6ecb4dd7c"
113
+ },
114
+ {
115
+ "dataPath": "params_shard_4.bin",
116
+ "format": "raw-shard",
117
+ "nbytes": 27525120,
118
+ "records": [
119
+ {
120
+ "name": "model.layers.1.mlp.down_proj.weight",
121
+ "shape": [
122
+ 1536,
123
+ 8960
124
+ ],
125
+ "dtype": "float16",
126
+ "format": "f32-to-bf16",
127
+ "nbytes": 27525120,
128
+ "byteOffset": 0
129
+ }
130
+ ],
131
+ "md5sum": "9decd2b72296182323d68b6233a3e47a"
132
+ },
133
+ {
134
+ "dataPath": "params_shard_5.bin",
135
+ "format": "raw-shard",
136
+ "nbytes": 55050240,
137
+ "records": [
138
+ {
139
+ "name": "model.layers.1.mlp.gate_up_proj.weight",
140
+ "shape": [
141
+ 17920,
142
+ 1536
143
+ ],
144
+ "dtype": "float16",
145
+ "format": "f32-to-bf16",
146
+ "nbytes": 55050240,
147
+ "byteOffset": 0
148
+ }
149
+ ],
150
+ "md5sum": "46f2860673f77add88bbe1e6d697be1b"
151
+ },
152
+ {
153
+ "dataPath": "params_shard_6.bin",
154
+ "format": "raw-shard",
155
+ "nbytes": 27525120,
156
+ "records": [
157
+ {
158
+ "name": "model.layers.10.mlp.down_proj.weight",
159
+ "shape": [
160
+ 1536,
161
+ 8960
162
+ ],
163
+ "dtype": "float16",
164
+ "format": "f32-to-bf16",
165
+ "nbytes": 27525120,
166
+ "byteOffset": 0
167
+ }
168
+ ],
169
+ "md5sum": "f260e42af4a1bc32aba36b7fd7d31116"
170
+ },
171
+ {
172
+ "dataPath": "params_shard_7.bin",
173
+ "format": "raw-shard",
174
+ "nbytes": 55050240,
175
+ "records": [
176
+ {
177
+ "name": "model.layers.10.mlp.gate_up_proj.weight",
178
+ "shape": [
179
+ 17920,
180
+ 1536
181
+ ],
182
+ "dtype": "float16",
183
+ "format": "f32-to-bf16",
184
+ "nbytes": 55050240,
185
+ "byteOffset": 0
186
+ }
187
+ ],
188
+ "md5sum": "2d3c8efd9e683084ef3bdc39ca2d65da"
189
+ },
190
+ {
191
+ "dataPath": "params_shard_8.bin",
192
+ "format": "raw-shard",
193
+ "nbytes": 27525120,
194
+ "records": [
195
+ {
196
+ "name": "model.layers.11.mlp.down_proj.weight",
197
+ "shape": [
198
+ 1536,
199
+ 8960
200
+ ],
201
+ "dtype": "float16",
202
+ "format": "f32-to-bf16",
203
+ "nbytes": 27525120,
204
+ "byteOffset": 0
205
+ }
206
+ ],
207
+ "md5sum": "4aebcfa9094ddc7eb84c1a0f7b8edcd5"
208
+ },
209
+ {
210
+ "dataPath": "params_shard_9.bin",
211
+ "format": "raw-shard",
212
+ "nbytes": 55050240,
213
+ "records": [
214
+ {
215
+ "name": "model.layers.11.mlp.gate_up_proj.weight",
216
+ "shape": [
217
+ 17920,
218
+ 1536
219
+ ],
220
+ "dtype": "float16",
221
+ "format": "f32-to-bf16",
222
+ "nbytes": 55050240,
223
+ "byteOffset": 0
224
+ }
225
+ ],
226
+ "md5sum": "28cd958ad4a9d0605066a9e1f2e14c0d"
227
+ },
228
+ {
229
+ "dataPath": "params_shard_10.bin",
230
+ "format": "raw-shard",
231
+ "nbytes": 33060864,
232
+ "records": [
233
+ {
234
+ "name": "model.layers.0.self_attn.c_attn.weight",
235
+ "shape": [
236
+ 2048,
237
+ 1536
238
+ ],
239
+ "dtype": "float16",
240
+ "format": "f32-to-bf16",
241
+ "nbytes": 6291456,
242
+ "byteOffset": 0
243
+ },
244
+ {
245
+ "name": "model.layers.0.self_attn.o_proj.weight",
246
+ "shape": [
247
+ 1536,
248
+ 1536
249
+ ],
250
+ "dtype": "float16",
251
+ "format": "f32-to-bf16",
252
+ "nbytes": 4718592,
253
+ "byteOffset": 6291456
254
+ },
255
+ {
256
+ "name": "model.layers.1.input_layernorm.weight",
257
+ "shape": [
258
+ 1536
259
+ ],
260
+ "dtype": "float16",
261
+ "format": "f32-to-bf16",
262
+ "nbytes": 3072,
263
+ "byteOffset": 11010048
264
+ },
265
+ {
266
+ "name": "model.layers.1.post_attention_layernorm.weight",
267
+ "shape": [
268
+ 1536
269
+ ],
270
+ "dtype": "float16",
271
+ "format": "f32-to-bf16",
272
+ "nbytes": 3072,
273
+ "byteOffset": 11013120
274
+ },
275
+ {
276
+ "name": "model.layers.1.self_attn.c_attn.bias",
277
+ "shape": [
278
+ 2048
279
+ ],
280
+ "dtype": "float16",
281
+ "format": "f32-to-bf16",
282
+ "nbytes": 4096,
283
+ "byteOffset": 11016192
284
+ },
285
+ {
286
+ "name": "model.layers.1.self_attn.c_attn.weight",
287
+ "shape": [
288
+ 2048,
289
+ 1536
290
+ ],
291
+ "dtype": "float16",
292
+ "format": "f32-to-bf16",
293
+ "nbytes": 6291456,
294
+ "byteOffset": 11020288
295
+ },
296
+ {
297
+ "name": "model.layers.1.self_attn.o_proj.weight",
298
+ "shape": [
299
+ 1536,
300
+ 1536
301
+ ],
302
+ "dtype": "float16",
303
+ "format": "f32-to-bf16",
304
+ "nbytes": 4718592,
305
+ "byteOffset": 17311744
306
+ },
307
+ {
308
+ "name": "model.layers.10.input_layernorm.weight",
309
+ "shape": [
310
+ 1536
311
+ ],
312
+ "dtype": "float16",
313
+ "format": "f32-to-bf16",
314
+ "nbytes": 3072,
315
+ "byteOffset": 22030336
316
+ },
317
+ {
318
+ "name": "model.layers.10.post_attention_layernorm.weight",
319
+ "shape": [
320
+ 1536
321
+ ],
322
+ "dtype": "float16",
323
+ "format": "f32-to-bf16",
324
+ "nbytes": 3072,
325
+ "byteOffset": 22033408
326
+ },
327
+ {
328
+ "name": "model.layers.10.self_attn.c_attn.bias",
329
+ "shape": [
330
+ 2048
331
+ ],
332
+ "dtype": "float16",
333
+ "format": "f32-to-bf16",
334
+ "nbytes": 4096,
335
+ "byteOffset": 22036480
336
+ },
337
+ {
338
+ "name": "model.layers.10.self_attn.c_attn.weight",
339
+ "shape": [
340
+ 2048,
341
+ 1536
342
+ ],
343
+ "dtype": "float16",
344
+ "format": "f32-to-bf16",
345
+ "nbytes": 6291456,
346
+ "byteOffset": 22040576
347
+ },
348
+ {
349
+ "name": "model.layers.10.self_attn.o_proj.weight",
350
+ "shape": [
351
+ 1536,
352
+ 1536
353
+ ],
354
+ "dtype": "float16",
355
+ "format": "f32-to-bf16",
356
+ "nbytes": 4718592,
357
+ "byteOffset": 28332032
358
+ },
359
+ {
360
+ "name": "model.layers.11.input_layernorm.weight",
361
+ "shape": [
362
+ 1536
363
+ ],
364
+ "dtype": "float16",
365
+ "format": "f32-to-bf16",
366
+ "nbytes": 3072,
367
+ "byteOffset": 33050624
368
+ },
369
+ {
370
+ "name": "model.layers.11.post_attention_layernorm.weight",
371
+ "shape": [
372
+ 1536
373
+ ],
374
+ "dtype": "float16",
375
+ "format": "f32-to-bf16",
376
+ "nbytes": 3072,
377
+ "byteOffset": 33053696
378
+ },
379
+ {
380
+ "name": "model.layers.11.self_attn.c_attn.bias",
381
+ "shape": [
382
+ 2048
383
+ ],
384
+ "dtype": "float16",
385
+ "format": "f32-to-bf16",
386
+ "nbytes": 4096,
387
+ "byteOffset": 33056768
388
+ }
389
+ ],
390
+ "md5sum": "724eafae7080ac9c64e2635b8f39dee1"
391
+ },
392
+ {
393
+ "dataPath": "params_shard_11.bin",
394
+ "format": "raw-shard",
395
+ "nbytes": 27525120,
396
+ "records": [
397
+ {
398
+ "name": "model.layers.12.mlp.down_proj.weight",
399
+ "shape": [
400
+ 1536,
401
+ 8960
402
+ ],
403
+ "dtype": "float16",
404
+ "format": "f32-to-bf16",
405
+ "nbytes": 27525120,
406
+ "byteOffset": 0
407
+ }
408
+ ],
409
+ "md5sum": "7c8fbc4469b959ca36ccb5b56b1aff7a"
410
+ },
411
+ {
412
+ "dataPath": "params_shard_12.bin",
413
+ "format": "raw-shard",
414
+ "nbytes": 55050240,
415
+ "records": [
416
+ {
417
+ "name": "model.layers.12.mlp.gate_up_proj.weight",
418
+ "shape": [
419
+ 17920,
420
+ 1536
421
+ ],
422
+ "dtype": "float16",
423
+ "format": "f32-to-bf16",
424
+ "nbytes": 55050240,
425
+ "byteOffset": 0
426
+ }
427
+ ],
428
+ "md5sum": "7dcc2b209c32558cf604f372054f94e1"
429
+ },
430
+ {
431
+ "dataPath": "params_shard_13.bin",
432
+ "format": "raw-shard",
433
+ "nbytes": 27525120,
434
+ "records": [
435
+ {
436
+ "name": "model.layers.13.mlp.down_proj.weight",
437
+ "shape": [
438
+ 1536,
439
+ 8960
440
+ ],
441
+ "dtype": "float16",
442
+ "format": "f32-to-bf16",
443
+ "nbytes": 27525120,
444
+ "byteOffset": 0
445
+ }
446
+ ],
447
+ "md5sum": "6c6d9014c54fc4be4c62c4d76bded576"
448
+ },
449
+ {
450
+ "dataPath": "params_shard_14.bin",
451
+ "format": "raw-shard",
452
+ "nbytes": 55050240,
453
+ "records": [
454
+ {
455
+ "name": "model.layers.13.mlp.gate_up_proj.weight",
456
+ "shape": [
457
+ 17920,
458
+ 1536
459
+ ],
460
+ "dtype": "float16",
461
+ "format": "f32-to-bf16",
462
+ "nbytes": 55050240,
463
+ "byteOffset": 0
464
+ }
465
+ ],
466
+ "md5sum": "783503713029d2e2e1e5ef970dbb4659"
467
+ },
468
+ {
469
+ "dataPath": "params_shard_15.bin",
470
+ "format": "raw-shard",
471
+ "nbytes": 27525120,
472
+ "records": [
473
+ {
474
+ "name": "model.layers.14.mlp.down_proj.weight",
475
+ "shape": [
476
+ 1536,
477
+ 8960
478
+ ],
479
+ "dtype": "float16",
480
+ "format": "f32-to-bf16",
481
+ "nbytes": 27525120,
482
+ "byteOffset": 0
483
+ }
484
+ ],
485
+ "md5sum": "3c3ad6b68589b515ff032534668446d7"
486
+ },
487
+ {
488
+ "dataPath": "params_shard_16.bin",
489
+ "format": "raw-shard",
490
+ "nbytes": 55050240,
491
+ "records": [
492
+ {
493
+ "name": "model.layers.14.mlp.gate_up_proj.weight",
494
+ "shape": [
495
+ 17920,
496
+ 1536
497
+ ],
498
+ "dtype": "float16",
499
+ "format": "f32-to-bf16",
500
+ "nbytes": 55050240,
501
+ "byteOffset": 0
502
+ }
503
+ ],
504
+ "md5sum": "829ef33937e95753fa00e48e71d26079"
505
+ },
506
+ {
507
+ "dataPath": "params_shard_17.bin",
508
+ "format": "raw-shard",
509
+ "nbytes": 33060864,
510
+ "records": [
511
+ {
512
+ "name": "model.layers.11.self_attn.c_attn.weight",
513
+ "shape": [
514
+ 2048,
515
+ 1536
516
+ ],
517
+ "dtype": "float16",
518
+ "format": "f32-to-bf16",
519
+ "nbytes": 6291456,
520
+ "byteOffset": 0
521
+ },
522
+ {
523
+ "name": "model.layers.11.self_attn.o_proj.weight",
524
+ "shape": [
525
+ 1536,
526
+ 1536
527
+ ],
528
+ "dtype": "float16",
529
+ "format": "f32-to-bf16",
530
+ "nbytes": 4718592,
531
+ "byteOffset": 6291456
532
+ },
533
+ {
534
+ "name": "model.layers.12.input_layernorm.weight",
535
+ "shape": [
536
+ 1536
537
+ ],
538
+ "dtype": "float16",
539
+ "format": "f32-to-bf16",
540
+ "nbytes": 3072,
541
+ "byteOffset": 11010048
542
+ },
543
+ {
544
+ "name": "model.layers.12.post_attention_layernorm.weight",
545
+ "shape": [
546
+ 1536
547
+ ],
548
+ "dtype": "float16",
549
+ "format": "f32-to-bf16",
550
+ "nbytes": 3072,
551
+ "byteOffset": 11013120
552
+ },
553
+ {
554
+ "name": "model.layers.12.self_attn.c_attn.bias",
555
+ "shape": [
556
+ 2048
557
+ ],
558
+ "dtype": "float16",
559
+ "format": "f32-to-bf16",
560
+ "nbytes": 4096,
561
+ "byteOffset": 11016192
562
+ },
563
+ {
564
+ "name": "model.layers.12.self_attn.c_attn.weight",
565
+ "shape": [
566
+ 2048,
567
+ 1536
568
+ ],
569
+ "dtype": "float16",
570
+ "format": "f32-to-bf16",
571
+ "nbytes": 6291456,
572
+ "byteOffset": 11020288
573
+ },
574
+ {
575
+ "name": "model.layers.12.self_attn.o_proj.weight",
576
+ "shape": [
577
+ 1536,
578
+ 1536
579
+ ],
580
+ "dtype": "float16",
581
+ "format": "f32-to-bf16",
582
+ "nbytes": 4718592,
583
+ "byteOffset": 17311744
584
+ },
585
+ {
586
+ "name": "model.layers.13.input_layernorm.weight",
587
+ "shape": [
588
+ 1536
589
+ ],
590
+ "dtype": "float16",
591
+ "format": "f32-to-bf16",
592
+ "nbytes": 3072,
593
+ "byteOffset": 22030336
594
+ },
595
+ {
596
+ "name": "model.layers.13.post_attention_layernorm.weight",
597
+ "shape": [
598
+ 1536
599
+ ],
600
+ "dtype": "float16",
601
+ "format": "f32-to-bf16",
602
+ "nbytes": 3072,
603
+ "byteOffset": 22033408
604
+ },
605
+ {
606
+ "name": "model.layers.13.self_attn.c_attn.bias",
607
+ "shape": [
608
+ 2048
609
+ ],
610
+ "dtype": "float16",
611
+ "format": "f32-to-bf16",
612
+ "nbytes": 4096,
613
+ "byteOffset": 22036480
614
+ },
615
+ {
616
+ "name": "model.layers.13.self_attn.c_attn.weight",
617
+ "shape": [
618
+ 2048,
619
+ 1536
620
+ ],
621
+ "dtype": "float16",
622
+ "format": "f32-to-bf16",
623
+ "nbytes": 6291456,
624
+ "byteOffset": 22040576
625
+ },
626
+ {
627
+ "name": "model.layers.13.self_attn.o_proj.weight",
628
+ "shape": [
629
+ 1536,
630
+ 1536
631
+ ],
632
+ "dtype": "float16",
633
+ "format": "f32-to-bf16",
634
+ "nbytes": 4718592,
635
+ "byteOffset": 28332032
636
+ },
637
+ {
638
+ "name": "model.layers.14.input_layernorm.weight",
639
+ "shape": [
640
+ 1536
641
+ ],
642
+ "dtype": "float16",
643
+ "format": "f32-to-bf16",
644
+ "nbytes": 3072,
645
+ "byteOffset": 33050624
646
+ },
647
+ {
648
+ "name": "model.layers.14.post_attention_layernorm.weight",
649
+ "shape": [
650
+ 1536
651
+ ],
652
+ "dtype": "float16",
653
+ "format": "f32-to-bf16",
654
+ "nbytes": 3072,
655
+ "byteOffset": 33053696
656
+ },
657
+ {
658
+ "name": "model.layers.14.self_attn.c_attn.bias",
659
+ "shape": [
660
+ 2048
661
+ ],
662
+ "dtype": "float16",
663
+ "format": "f32-to-bf16",
664
+ "nbytes": 4096,
665
+ "byteOffset": 33056768
666
+ }
667
+ ],
668
+ "md5sum": "1011c7025b617580675ce288695c9320"
669
+ },
670
+ {
671
+ "dataPath": "params_shard_18.bin",
672
+ "format": "raw-shard",
673
+ "nbytes": 27525120,
674
+ "records": [
675
+ {
676
+ "name": "model.layers.15.mlp.down_proj.weight",
677
+ "shape": [
678
+ 1536,
679
+ 8960
680
+ ],
681
+ "dtype": "float16",
682
+ "format": "f32-to-bf16",
683
+ "nbytes": 27525120,
684
+ "byteOffset": 0
685
+ }
686
+ ],
687
+ "md5sum": "1ec33da19abd8946ec3ce6eff05a00f2"
688
+ },
689
+ {
690
+ "dataPath": "params_shard_19.bin",
691
+ "format": "raw-shard",
692
+ "nbytes": 55050240,
693
+ "records": [
694
+ {
695
+ "name": "model.layers.15.mlp.gate_up_proj.weight",
696
+ "shape": [
697
+ 17920,
698
+ 1536
699
+ ],
700
+ "dtype": "float16",
701
+ "format": "f32-to-bf16",
702
+ "nbytes": 55050240,
703
+ "byteOffset": 0
704
+ }
705
+ ],
706
+ "md5sum": "80bada2591e989c2a2d893ac7ca17d92"
707
+ },
708
+ {
709
+ "dataPath": "params_shard_20.bin",
710
+ "format": "raw-shard",
711
+ "nbytes": 27525120,
712
+ "records": [
713
+ {
714
+ "name": "model.layers.16.mlp.down_proj.weight",
715
+ "shape": [
716
+ 1536,
717
+ 8960
718
+ ],
719
+ "dtype": "float16",
720
+ "format": "f32-to-bf16",
721
+ "nbytes": 27525120,
722
+ "byteOffset": 0
723
+ }
724
+ ],
725
+ "md5sum": "bc2cfd054525906601b86dfaef89f799"
726
+ },
727
+ {
728
+ "dataPath": "params_shard_21.bin",
729
+ "format": "raw-shard",
730
+ "nbytes": 55050240,
731
+ "records": [
732
+ {
733
+ "name": "model.layers.16.mlp.gate_up_proj.weight",
734
+ "shape": [
735
+ 17920,
736
+ 1536
737
+ ],
738
+ "dtype": "float16",
739
+ "format": "f32-to-bf16",
740
+ "nbytes": 55050240,
741
+ "byteOffset": 0
742
+ }
743
+ ],
744
+ "md5sum": "689386905f4503dc078865cead2e2516"
745
+ },
746
+ {
747
+ "dataPath": "params_shard_22.bin",
748
+ "format": "raw-shard",
749
+ "nbytes": 27525120,
750
+ "records": [
751
+ {
752
+ "name": "model.layers.17.mlp.down_proj.weight",
753
+ "shape": [
754
+ 1536,
755
+ 8960
756
+ ],
757
+ "dtype": "float16",
758
+ "format": "f32-to-bf16",
759
+ "nbytes": 27525120,
760
+ "byteOffset": 0
761
+ }
762
+ ],
763
+ "md5sum": "9f074861b1282a53d624aa8271eb7ceb"
764
+ },
765
+ {
766
+ "dataPath": "params_shard_23.bin",
767
+ "format": "raw-shard",
768
+ "nbytes": 55050240,
769
+ "records": [
770
+ {
771
+ "name": "model.layers.17.mlp.gate_up_proj.weight",
772
+ "shape": [
773
+ 17920,
774
+ 1536
775
+ ],
776
+ "dtype": "float16",
777
+ "format": "f32-to-bf16",
778
+ "nbytes": 55050240,
779
+ "byteOffset": 0
780
+ }
781
+ ],
782
+ "md5sum": "2151c0b55962940f406023b7abb50ced"
783
+ },
784
+ {
785
+ "dataPath": "params_shard_24.bin",
786
+ "format": "raw-shard",
787
+ "nbytes": 33060864,
788
+ "records": [
789
+ {
790
+ "name": "model.layers.14.self_attn.c_attn.weight",
791
+ "shape": [
792
+ 2048,
793
+ 1536
794
+ ],
795
+ "dtype": "float16",
796
+ "format": "f32-to-bf16",
797
+ "nbytes": 6291456,
798
+ "byteOffset": 0
799
+ },
800
+ {
801
+ "name": "model.layers.14.self_attn.o_proj.weight",
802
+ "shape": [
803
+ 1536,
804
+ 1536
805
+ ],
806
+ "dtype": "float16",
807
+ "format": "f32-to-bf16",
808
+ "nbytes": 4718592,
809
+ "byteOffset": 6291456
810
+ },
811
+ {
812
+ "name": "model.layers.15.input_layernorm.weight",
813
+ "shape": [
814
+ 1536
815
+ ],
816
+ "dtype": "float16",
817
+ "format": "f32-to-bf16",
818
+ "nbytes": 3072,
819
+ "byteOffset": 11010048
820
+ },
821
+ {
822
+ "name": "model.layers.15.post_attention_layernorm.weight",
823
+ "shape": [
824
+ 1536
825
+ ],
826
+ "dtype": "float16",
827
+ "format": "f32-to-bf16",
828
+ "nbytes": 3072,
829
+ "byteOffset": 11013120
830
+ },
831
+ {
832
+ "name": "model.layers.15.self_attn.c_attn.bias",
833
+ "shape": [
834
+ 2048
835
+ ],
836
+ "dtype": "float16",
837
+ "format": "f32-to-bf16",
838
+ "nbytes": 4096,
839
+ "byteOffset": 11016192
840
+ },
841
+ {
842
+ "name": "model.layers.15.self_attn.c_attn.weight",
843
+ "shape": [
844
+ 2048,
845
+ 1536
846
+ ],
847
+ "dtype": "float16",
848
+ "format": "f32-to-bf16",
849
+ "nbytes": 6291456,
850
+ "byteOffset": 11020288
851
+ },
852
+ {
853
+ "name": "model.layers.15.self_attn.o_proj.weight",
854
+ "shape": [
855
+ 1536,
856
+ 1536
857
+ ],
858
+ "dtype": "float16",
859
+ "format": "f32-to-bf16",
860
+ "nbytes": 4718592,
861
+ "byteOffset": 17311744
862
+ },
863
+ {
864
+ "name": "model.layers.16.input_layernorm.weight",
865
+ "shape": [
866
+ 1536
867
+ ],
868
+ "dtype": "float16",
869
+ "format": "f32-to-bf16",
870
+ "nbytes": 3072,
871
+ "byteOffset": 22030336
872
+ },
873
+ {
874
+ "name": "model.layers.16.post_attention_layernorm.weight",
875
+ "shape": [
876
+ 1536
877
+ ],
878
+ "dtype": "float16",
879
+ "format": "f32-to-bf16",
880
+ "nbytes": 3072,
881
+ "byteOffset": 22033408
882
+ },
883
+ {
884
+ "name": "model.layers.16.self_attn.c_attn.bias",
885
+ "shape": [
886
+ 2048
887
+ ],
888
+ "dtype": "float16",
889
+ "format": "f32-to-bf16",
890
+ "nbytes": 4096,
891
+ "byteOffset": 22036480
892
+ },
893
+ {
894
+ "name": "model.layers.16.self_attn.c_attn.weight",
895
+ "shape": [
896
+ 2048,
897
+ 1536
898
+ ],
899
+ "dtype": "float16",
900
+ "format": "f32-to-bf16",
901
+ "nbytes": 6291456,
902
+ "byteOffset": 22040576
903
+ },
904
+ {
905
+ "name": "model.layers.16.self_attn.o_proj.weight",
906
+ "shape": [
907
+ 1536,
908
+ 1536
909
+ ],
910
+ "dtype": "float16",
911
+ "format": "f32-to-bf16",
912
+ "nbytes": 4718592,
913
+ "byteOffset": 28332032
914
+ },
915
+ {
916
+ "name": "model.layers.17.input_layernorm.weight",
917
+ "shape": [
918
+ 1536
919
+ ],
920
+ "dtype": "float16",
921
+ "format": "f32-to-bf16",
922
+ "nbytes": 3072,
923
+ "byteOffset": 33050624
924
+ },
925
+ {
926
+ "name": "model.layers.17.post_attention_layernorm.weight",
927
+ "shape": [
928
+ 1536
929
+ ],
930
+ "dtype": "float16",
931
+ "format": "f32-to-bf16",
932
+ "nbytes": 3072,
933
+ "byteOffset": 33053696
934
+ },
935
+ {
936
+ "name": "model.layers.17.self_attn.c_attn.bias",
937
+ "shape": [
938
+ 2048
939
+ ],
940
+ "dtype": "float16",
941
+ "format": "f32-to-bf16",
942
+ "nbytes": 4096,
943
+ "byteOffset": 33056768
944
+ }
945
+ ],
946
+ "md5sum": "e0f33779f14209c649609770249e910e"
947
+ },
948
+ {
949
+ "dataPath": "params_shard_25.bin",
950
+ "format": "raw-shard",
951
+ "nbytes": 27525120,
952
+ "records": [
953
+ {
954
+ "name": "model.layers.18.mlp.down_proj.weight",
955
+ "shape": [
956
+ 1536,
957
+ 8960
958
+ ],
959
+ "dtype": "float16",
960
+ "format": "f32-to-bf16",
961
+ "nbytes": 27525120,
962
+ "byteOffset": 0
963
+ }
964
+ ],
965
+ "md5sum": "6be106e14a60e9192d92b9bc5e03ab63"
966
+ },
967
+ {
968
+ "dataPath": "params_shard_26.bin",
969
+ "format": "raw-shard",
970
+ "nbytes": 55050240,
971
+ "records": [
972
+ {
973
+ "name": "model.layers.18.mlp.gate_up_proj.weight",
974
+ "shape": [
975
+ 17920,
976
+ 1536
977
+ ],
978
+ "dtype": "float16",
979
+ "format": "f32-to-bf16",
980
+ "nbytes": 55050240,
981
+ "byteOffset": 0
982
+ }
983
+ ],
984
+ "md5sum": "730cee18923347893bbb44b3653dcd4c"
985
+ },
986
+ {
987
+ "dataPath": "params_shard_27.bin",
988
+ "format": "raw-shard",
989
+ "nbytes": 27525120,
990
+ "records": [
991
+ {
992
+ "name": "model.layers.19.mlp.down_proj.weight",
993
+ "shape": [
994
+ 1536,
995
+ 8960
996
+ ],
997
+ "dtype": "float16",
998
+ "format": "f32-to-bf16",
999
+ "nbytes": 27525120,
1000
+ "byteOffset": 0
1001
+ }
1002
+ ],
1003
+ "md5sum": "68ac956faf4373deffce9e701b13312e"
1004
+ },
1005
+ {
1006
+ "dataPath": "params_shard_28.bin",
1007
+ "format": "raw-shard",
1008
+ "nbytes": 55050240,
1009
+ "records": [
1010
+ {
1011
+ "name": "model.layers.19.mlp.gate_up_proj.weight",
1012
+ "shape": [
1013
+ 17920,
1014
+ 1536
1015
+ ],
1016
+ "dtype": "float16",
1017
+ "format": "f32-to-bf16",
1018
+ "nbytes": 55050240,
1019
+ "byteOffset": 0
1020
+ }
1021
+ ],
1022
+ "md5sum": "13cfe0b30478bdb5907d2f97062b96a6"
1023
+ },
1024
+ {
1025
+ "dataPath": "params_shard_29.bin",
1026
+ "format": "raw-shard",
1027
+ "nbytes": 27525120,
1028
+ "records": [
1029
+ {
1030
+ "name": "model.layers.2.mlp.down_proj.weight",
1031
+ "shape": [
1032
+ 1536,
1033
+ 8960
1034
+ ],
1035
+ "dtype": "float16",
1036
+ "format": "f32-to-bf16",
1037
+ "nbytes": 27525120,
1038
+ "byteOffset": 0
1039
+ }
1040
+ ],
1041
+ "md5sum": "d22b27cf62cf897d4afc59b56df26dfd"
1042
+ },
1043
+ {
1044
+ "dataPath": "params_shard_30.bin",
1045
+ "format": "raw-shard",
1046
+ "nbytes": 55050240,
1047
+ "records": [
1048
+ {
1049
+ "name": "model.layers.2.mlp.gate_up_proj.weight",
1050
+ "shape": [
1051
+ 17920,
1052
+ 1536
1053
+ ],
1054
+ "dtype": "float16",
1055
+ "format": "f32-to-bf16",
1056
+ "nbytes": 55050240,
1057
+ "byteOffset": 0
1058
+ }
1059
+ ],
1060
+ "md5sum": "576f048acbf3d1be4814d7d5c8924ccf"
1061
+ },
1062
+ {
1063
+ "dataPath": "params_shard_31.bin",
1064
+ "format": "raw-shard",
1065
+ "nbytes": 33060864,
1066
+ "records": [
1067
+ {
1068
+ "name": "model.layers.17.self_attn.c_attn.weight",
1069
+ "shape": [
1070
+ 2048,
1071
+ 1536
1072
+ ],
1073
+ "dtype": "float16",
1074
+ "format": "f32-to-bf16",
1075
+ "nbytes": 6291456,
1076
+ "byteOffset": 0
1077
+ },
1078
+ {
1079
+ "name": "model.layers.17.self_attn.o_proj.weight",
1080
+ "shape": [
1081
+ 1536,
1082
+ 1536
1083
+ ],
1084
+ "dtype": "float16",
1085
+ "format": "f32-to-bf16",
1086
+ "nbytes": 4718592,
1087
+ "byteOffset": 6291456
1088
+ },
1089
+ {
1090
+ "name": "model.layers.18.input_layernorm.weight",
1091
+ "shape": [
1092
+ 1536
1093
+ ],
1094
+ "dtype": "float16",
1095
+ "format": "f32-to-bf16",
1096
+ "nbytes": 3072,
1097
+ "byteOffset": 11010048
1098
+ },
1099
+ {
1100
+ "name": "model.layers.18.post_attention_layernorm.weight",
1101
+ "shape": [
1102
+ 1536
1103
+ ],
1104
+ "dtype": "float16",
1105
+ "format": "f32-to-bf16",
1106
+ "nbytes": 3072,
1107
+ "byteOffset": 11013120
1108
+ },
1109
+ {
1110
+ "name": "model.layers.18.self_attn.c_attn.bias",
1111
+ "shape": [
1112
+ 2048
1113
+ ],
1114
+ "dtype": "float16",
1115
+ "format": "f32-to-bf16",
1116
+ "nbytes": 4096,
1117
+ "byteOffset": 11016192
1118
+ },
1119
+ {
1120
+ "name": "model.layers.18.self_attn.c_attn.weight",
1121
+ "shape": [
1122
+ 2048,
1123
+ 1536
1124
+ ],
1125
+ "dtype": "float16",
1126
+ "format": "f32-to-bf16",
1127
+ "nbytes": 6291456,
1128
+ "byteOffset": 11020288
1129
+ },
1130
+ {
1131
+ "name": "model.layers.18.self_attn.o_proj.weight",
1132
+ "shape": [
1133
+ 1536,
1134
+ 1536
1135
+ ],
1136
+ "dtype": "float16",
1137
+ "format": "f32-to-bf16",
1138
+ "nbytes": 4718592,
1139
+ "byteOffset": 17311744
1140
+ },
1141
+ {
1142
+ "name": "model.layers.19.input_layernorm.weight",
1143
+ "shape": [
1144
+ 1536
1145
+ ],
1146
+ "dtype": "float16",
1147
+ "format": "f32-to-bf16",
1148
+ "nbytes": 3072,
1149
+ "byteOffset": 22030336
1150
+ },
1151
+ {
1152
+ "name": "model.layers.19.post_attention_layernorm.weight",
1153
+ "shape": [
1154
+ 1536
1155
+ ],
1156
+ "dtype": "float16",
1157
+ "format": "f32-to-bf16",
1158
+ "nbytes": 3072,
1159
+ "byteOffset": 22033408
1160
+ },
1161
+ {
1162
+ "name": "model.layers.19.self_attn.c_attn.bias",
1163
+ "shape": [
1164
+ 2048
1165
+ ],
1166
+ "dtype": "float16",
1167
+ "format": "f32-to-bf16",
1168
+ "nbytes": 4096,
1169
+ "byteOffset": 22036480
1170
+ },
1171
+ {
1172
+ "name": "model.layers.19.self_attn.c_attn.weight",
1173
+ "shape": [
1174
+ 2048,
1175
+ 1536
1176
+ ],
1177
+ "dtype": "float16",
1178
+ "format": "f32-to-bf16",
1179
+ "nbytes": 6291456,
1180
+ "byteOffset": 22040576
1181
+ },
1182
+ {
1183
+ "name": "model.layers.19.self_attn.o_proj.weight",
1184
+ "shape": [
1185
+ 1536,
1186
+ 1536
1187
+ ],
1188
+ "dtype": "float16",
1189
+ "format": "f32-to-bf16",
1190
+ "nbytes": 4718592,
1191
+ "byteOffset": 28332032
1192
+ },
1193
+ {
1194
+ "name": "model.layers.2.input_layernorm.weight",
1195
+ "shape": [
1196
+ 1536
1197
+ ],
1198
+ "dtype": "float16",
1199
+ "format": "f32-to-bf16",
1200
+ "nbytes": 3072,
1201
+ "byteOffset": 33050624
1202
+ },
1203
+ {
1204
+ "name": "model.layers.2.post_attention_layernorm.weight",
1205
+ "shape": [
1206
+ 1536
1207
+ ],
1208
+ "dtype": "float16",
1209
+ "format": "f32-to-bf16",
1210
+ "nbytes": 3072,
1211
+ "byteOffset": 33053696
1212
+ },
1213
+ {
1214
+ "name": "model.layers.2.self_attn.c_attn.bias",
1215
+ "shape": [
1216
+ 2048
1217
+ ],
1218
+ "dtype": "float16",
1219
+ "format": "f32-to-bf16",
1220
+ "nbytes": 4096,
1221
+ "byteOffset": 33056768
1222
+ }
1223
+ ],
1224
+ "md5sum": "ba35ae2d08939bf3a86b5634f1c2fc50"
1225
+ },
1226
+ {
1227
+ "dataPath": "params_shard_32.bin",
1228
+ "format": "raw-shard",
1229
+ "nbytes": 27525120,
1230
+ "records": [
1231
+ {
1232
+ "name": "model.layers.20.mlp.down_proj.weight",
1233
+ "shape": [
1234
+ 1536,
1235
+ 8960
1236
+ ],
1237
+ "dtype": "float16",
1238
+ "format": "f32-to-bf16",
1239
+ "nbytes": 27525120,
1240
+ "byteOffset": 0
1241
+ }
1242
+ ],
1243
+ "md5sum": "cea4bfe4d20533262bbfd1a5f55414a0"
1244
+ },
1245
+ {
1246
+ "dataPath": "params_shard_33.bin",
1247
+ "format": "raw-shard",
1248
+ "nbytes": 55050240,
1249
+ "records": [
1250
+ {
1251
+ "name": "model.layers.20.mlp.gate_up_proj.weight",
1252
+ "shape": [
1253
+ 17920,
1254
+ 1536
1255
+ ],
1256
+ "dtype": "float16",
1257
+ "format": "f32-to-bf16",
1258
+ "nbytes": 55050240,
1259
+ "byteOffset": 0
1260
+ }
1261
+ ],
1262
+ "md5sum": "3bb671f6852a83ace483b8774e4702c1"
1263
+ },
1264
+ {
1265
+ "dataPath": "params_shard_34.bin",
1266
+ "format": "raw-shard",
1267
+ "nbytes": 27525120,
1268
+ "records": [
1269
+ {
1270
+ "name": "model.layers.21.mlp.down_proj.weight",
1271
+ "shape": [
1272
+ 1536,
1273
+ 8960
1274
+ ],
1275
+ "dtype": "float16",
1276
+ "format": "f32-to-bf16",
1277
+ "nbytes": 27525120,
1278
+ "byteOffset": 0
1279
+ }
1280
+ ],
1281
+ "md5sum": "a1454ea96781f0988c17c5103285e57e"
1282
+ },
1283
+ {
1284
+ "dataPath": "params_shard_35.bin",
1285
+ "format": "raw-shard",
1286
+ "nbytes": 55050240,
1287
+ "records": [
1288
+ {
1289
+ "name": "model.layers.21.mlp.gate_up_proj.weight",
1290
+ "shape": [
1291
+ 17920,
1292
+ 1536
1293
+ ],
1294
+ "dtype": "float16",
1295
+ "format": "f32-to-bf16",
1296
+ "nbytes": 55050240,
1297
+ "byteOffset": 0
1298
+ }
1299
+ ],
1300
+ "md5sum": "8a90e1a3f9a3205d686662193403b96f"
1301
+ },
1302
+ {
1303
+ "dataPath": "params_shard_36.bin",
1304
+ "format": "raw-shard",
1305
+ "nbytes": 27525120,
1306
+ "records": [
1307
+ {
1308
+ "name": "model.layers.22.mlp.down_proj.weight",
1309
+ "shape": [
1310
+ 1536,
1311
+ 8960
1312
+ ],
1313
+ "dtype": "float16",
1314
+ "format": "f32-to-bf16",
1315
+ "nbytes": 27525120,
1316
+ "byteOffset": 0
1317
+ }
1318
+ ],
1319
+ "md5sum": "d5ccdd836cd755e7b75ca6c57d52349f"
1320
+ },
1321
+ {
1322
+ "dataPath": "params_shard_37.bin",
1323
+ "format": "raw-shard",
1324
+ "nbytes": 55050240,
1325
+ "records": [
1326
+ {
1327
+ "name": "model.layers.22.mlp.gate_up_proj.weight",
1328
+ "shape": [
1329
+ 17920,
1330
+ 1536
1331
+ ],
1332
+ "dtype": "float16",
1333
+ "format": "f32-to-bf16",
1334
+ "nbytes": 55050240,
1335
+ "byteOffset": 0
1336
+ }
1337
+ ],
1338
+ "md5sum": "4c16ce0ea740f7c4cfe53c2d767c0932"
1339
+ },
1340
+ {
1341
+ "dataPath": "params_shard_38.bin",
1342
+ "format": "raw-shard",
1343
+ "nbytes": 33060864,
1344
+ "records": [
1345
+ {
1346
+ "name": "model.layers.2.self_attn.c_attn.weight",
1347
+ "shape": [
1348
+ 2048,
1349
+ 1536
1350
+ ],
1351
+ "dtype": "float16",
1352
+ "format": "f32-to-bf16",
1353
+ "nbytes": 6291456,
1354
+ "byteOffset": 0
1355
+ },
1356
+ {
1357
+ "name": "model.layers.2.self_attn.o_proj.weight",
1358
+ "shape": [
1359
+ 1536,
1360
+ 1536
1361
+ ],
1362
+ "dtype": "float16",
1363
+ "format": "f32-to-bf16",
1364
+ "nbytes": 4718592,
1365
+ "byteOffset": 6291456
1366
+ },
1367
+ {
1368
+ "name": "model.layers.20.input_layernorm.weight",
1369
+ "shape": [
1370
+ 1536
1371
+ ],
1372
+ "dtype": "float16",
1373
+ "format": "f32-to-bf16",
1374
+ "nbytes": 3072,
1375
+ "byteOffset": 11010048
1376
+ },
1377
+ {
1378
+ "name": "model.layers.20.post_attention_layernorm.weight",
1379
+ "shape": [
1380
+ 1536
1381
+ ],
1382
+ "dtype": "float16",
1383
+ "format": "f32-to-bf16",
1384
+ "nbytes": 3072,
1385
+ "byteOffset": 11013120
1386
+ },
1387
+ {
1388
+ "name": "model.layers.20.self_attn.c_attn.bias",
1389
+ "shape": [
1390
+ 2048
1391
+ ],
1392
+ "dtype": "float16",
1393
+ "format": "f32-to-bf16",
1394
+ "nbytes": 4096,
1395
+ "byteOffset": 11016192
1396
+ },
1397
+ {
1398
+ "name": "model.layers.20.self_attn.c_attn.weight",
1399
+ "shape": [
1400
+ 2048,
1401
+ 1536
1402
+ ],
1403
+ "dtype": "float16",
1404
+ "format": "f32-to-bf16",
1405
+ "nbytes": 6291456,
1406
+ "byteOffset": 11020288
1407
+ },
1408
+ {
1409
+ "name": "model.layers.20.self_attn.o_proj.weight",
1410
+ "shape": [
1411
+ 1536,
1412
+ 1536
1413
+ ],
1414
+ "dtype": "float16",
1415
+ "format": "f32-to-bf16",
1416
+ "nbytes": 4718592,
1417
+ "byteOffset": 17311744
1418
+ },
1419
+ {
1420
+ "name": "model.layers.21.input_layernorm.weight",
1421
+ "shape": [
1422
+ 1536
1423
+ ],
1424
+ "dtype": "float16",
1425
+ "format": "f32-to-bf16",
1426
+ "nbytes": 3072,
1427
+ "byteOffset": 22030336
1428
+ },
1429
+ {
1430
+ "name": "model.layers.21.post_attention_layernorm.weight",
1431
+ "shape": [
1432
+ 1536
1433
+ ],
1434
+ "dtype": "float16",
1435
+ "format": "f32-to-bf16",
1436
+ "nbytes": 3072,
1437
+ "byteOffset": 22033408
1438
+ },
1439
+ {
1440
+ "name": "model.layers.21.self_attn.c_attn.bias",
1441
+ "shape": [
1442
+ 2048
1443
+ ],
1444
+ "dtype": "float16",
1445
+ "format": "f32-to-bf16",
1446
+ "nbytes": 4096,
1447
+ "byteOffset": 22036480
1448
+ },
1449
+ {
1450
+ "name": "model.layers.21.self_attn.c_attn.weight",
1451
+ "shape": [
1452
+ 2048,
1453
+ 1536
1454
+ ],
1455
+ "dtype": "float16",
1456
+ "format": "f32-to-bf16",
1457
+ "nbytes": 6291456,
1458
+ "byteOffset": 22040576
1459
+ },
1460
+ {
1461
+ "name": "model.layers.21.self_attn.o_proj.weight",
1462
+ "shape": [
1463
+ 1536,
1464
+ 1536
1465
+ ],
1466
+ "dtype": "float16",
1467
+ "format": "f32-to-bf16",
1468
+ "nbytes": 4718592,
1469
+ "byteOffset": 28332032
1470
+ },
1471
+ {
1472
+ "name": "model.layers.22.input_layernorm.weight",
1473
+ "shape": [
1474
+ 1536
1475
+ ],
1476
+ "dtype": "float16",
1477
+ "format": "f32-to-bf16",
1478
+ "nbytes": 3072,
1479
+ "byteOffset": 33050624
1480
+ },
1481
+ {
1482
+ "name": "model.layers.22.post_attention_layernorm.weight",
1483
+ "shape": [
1484
+ 1536
1485
+ ],
1486
+ "dtype": "float16",
1487
+ "format": "f32-to-bf16",
1488
+ "nbytes": 3072,
1489
+ "byteOffset": 33053696
1490
+ },
1491
+ {
1492
+ "name": "model.layers.22.self_attn.c_attn.bias",
1493
+ "shape": [
1494
+ 2048
1495
+ ],
1496
+ "dtype": "float16",
1497
+ "format": "f32-to-bf16",
1498
+ "nbytes": 4096,
1499
+ "byteOffset": 33056768
1500
+ }
1501
+ ],
1502
+ "md5sum": "f46d7d3c3f1f8ee51a5091a056e7469f"
1503
+ },
1504
+ {
1505
+ "dataPath": "params_shard_39.bin",
1506
+ "format": "raw-shard",
1507
+ "nbytes": 27525120,
1508
+ "records": [
1509
+ {
1510
+ "name": "model.layers.23.mlp.down_proj.weight",
1511
+ "shape": [
1512
+ 1536,
1513
+ 8960
1514
+ ],
1515
+ "dtype": "float16",
1516
+ "format": "f32-to-bf16",
1517
+ "nbytes": 27525120,
1518
+ "byteOffset": 0
1519
+ }
1520
+ ],
1521
+ "md5sum": "67ba46d9daf2ea224cdce3df4bbee33e"
1522
+ },
1523
+ {
1524
+ "dataPath": "params_shard_40.bin",
1525
+ "format": "raw-shard",
1526
+ "nbytes": 55050240,
1527
+ "records": [
1528
+ {
1529
+ "name": "model.layers.23.mlp.gate_up_proj.weight",
1530
+ "shape": [
1531
+ 17920,
1532
+ 1536
1533
+ ],
1534
+ "dtype": "float16",
1535
+ "format": "f32-to-bf16",
1536
+ "nbytes": 55050240,
1537
+ "byteOffset": 0
1538
+ }
1539
+ ],
1540
+ "md5sum": "c8b688c31f6987ccf6fe43b600ec578c"
1541
+ },
1542
+ {
1543
+ "dataPath": "params_shard_41.bin",
1544
+ "format": "raw-shard",
1545
+ "nbytes": 27525120,
1546
+ "records": [
1547
+ {
1548
+ "name": "model.layers.24.mlp.down_proj.weight",
1549
+ "shape": [
1550
+ 1536,
1551
+ 8960
1552
+ ],
1553
+ "dtype": "float16",
1554
+ "format": "f32-to-bf16",
1555
+ "nbytes": 27525120,
1556
+ "byteOffset": 0
1557
+ }
1558
+ ],
1559
+ "md5sum": "43865758cc6d7f7ce6cca424e1294f6a"
1560
+ },
1561
+ {
1562
+ "dataPath": "params_shard_42.bin",
1563
+ "format": "raw-shard",
1564
+ "nbytes": 55050240,
1565
+ "records": [
1566
+ {
1567
+ "name": "model.layers.24.mlp.gate_up_proj.weight",
1568
+ "shape": [
1569
+ 17920,
1570
+ 1536
1571
+ ],
1572
+ "dtype": "float16",
1573
+ "format": "f32-to-bf16",
1574
+ "nbytes": 55050240,
1575
+ "byteOffset": 0
1576
+ }
1577
+ ],
1578
+ "md5sum": "b344e1c3df1d97f91fac114b322f6d8a"
1579
+ },
1580
+ {
1581
+ "dataPath": "params_shard_43.bin",
1582
+ "format": "raw-shard",
1583
+ "nbytes": 27525120,
1584
+ "records": [
1585
+ {
1586
+ "name": "model.layers.25.mlp.down_proj.weight",
1587
+ "shape": [
1588
+ 1536,
1589
+ 8960
1590
+ ],
1591
+ "dtype": "float16",
1592
+ "format": "f32-to-bf16",
1593
+ "nbytes": 27525120,
1594
+ "byteOffset": 0
1595
+ }
1596
+ ],
1597
+ "md5sum": "d257c98f83e522584fe43139c4e54356"
1598
+ },
1599
+ {
1600
+ "dataPath": "params_shard_44.bin",
1601
+ "format": "raw-shard",
1602
+ "nbytes": 55050240,
1603
+ "records": [
1604
+ {
1605
+ "name": "model.layers.25.mlp.gate_up_proj.weight",
1606
+ "shape": [
1607
+ 17920,
1608
+ 1536
1609
+ ],
1610
+ "dtype": "float16",
1611
+ "format": "f32-to-bf16",
1612
+ "nbytes": 55050240,
1613
+ "byteOffset": 0
1614
+ }
1615
+ ],
1616
+ "md5sum": "c2979b3dcaf44c5ab44ce23603a59710"
1617
+ },
1618
+ {
1619
+ "dataPath": "params_shard_45.bin",
1620
+ "format": "raw-shard",
1621
+ "nbytes": 33060864,
1622
+ "records": [
1623
+ {
1624
+ "name": "model.layers.22.self_attn.c_attn.weight",
1625
+ "shape": [
1626
+ 2048,
1627
+ 1536
1628
+ ],
1629
+ "dtype": "float16",
1630
+ "format": "f32-to-bf16",
1631
+ "nbytes": 6291456,
1632
+ "byteOffset": 0
1633
+ },
1634
+ {
1635
+ "name": "model.layers.22.self_attn.o_proj.weight",
1636
+ "shape": [
1637
+ 1536,
1638
+ 1536
1639
+ ],
1640
+ "dtype": "float16",
1641
+ "format": "f32-to-bf16",
1642
+ "nbytes": 4718592,
1643
+ "byteOffset": 6291456
1644
+ },
1645
+ {
1646
+ "name": "model.layers.23.input_layernorm.weight",
1647
+ "shape": [
1648
+ 1536
1649
+ ],
1650
+ "dtype": "float16",
1651
+ "format": "f32-to-bf16",
1652
+ "nbytes": 3072,
1653
+ "byteOffset": 11010048
1654
+ },
1655
+ {
1656
+ "name": "model.layers.23.post_attention_layernorm.weight",
1657
+ "shape": [
1658
+ 1536
1659
+ ],
1660
+ "dtype": "float16",
1661
+ "format": "f32-to-bf16",
1662
+ "nbytes": 3072,
1663
+ "byteOffset": 11013120
1664
+ },
1665
+ {
1666
+ "name": "model.layers.23.self_attn.c_attn.bias",
1667
+ "shape": [
1668
+ 2048
1669
+ ],
1670
+ "dtype": "float16",
1671
+ "format": "f32-to-bf16",
1672
+ "nbytes": 4096,
1673
+ "byteOffset": 11016192
1674
+ },
1675
+ {
1676
+ "name": "model.layers.23.self_attn.c_attn.weight",
1677
+ "shape": [
1678
+ 2048,
1679
+ 1536
1680
+ ],
1681
+ "dtype": "float16",
1682
+ "format": "f32-to-bf16",
1683
+ "nbytes": 6291456,
1684
+ "byteOffset": 11020288
1685
+ },
1686
+ {
1687
+ "name": "model.layers.23.self_attn.o_proj.weight",
1688
+ "shape": [
1689
+ 1536,
1690
+ 1536
1691
+ ],
1692
+ "dtype": "float16",
1693
+ "format": "f32-to-bf16",
1694
+ "nbytes": 4718592,
1695
+ "byteOffset": 17311744
1696
+ },
1697
+ {
1698
+ "name": "model.layers.24.input_layernorm.weight",
1699
+ "shape": [
1700
+ 1536
1701
+ ],
1702
+ "dtype": "float16",
1703
+ "format": "f32-to-bf16",
1704
+ "nbytes": 3072,
1705
+ "byteOffset": 22030336
1706
+ },
1707
+ {
1708
+ "name": "model.layers.24.post_attention_layernorm.weight",
1709
+ "shape": [
1710
+ 1536
1711
+ ],
1712
+ "dtype": "float16",
1713
+ "format": "f32-to-bf16",
1714
+ "nbytes": 3072,
1715
+ "byteOffset": 22033408
1716
+ },
1717
+ {
1718
+ "name": "model.layers.24.self_attn.c_attn.bias",
1719
+ "shape": [
1720
+ 2048
1721
+ ],
1722
+ "dtype": "float16",
1723
+ "format": "f32-to-bf16",
1724
+ "nbytes": 4096,
1725
+ "byteOffset": 22036480
1726
+ },
1727
+ {
1728
+ "name": "model.layers.24.self_attn.c_attn.weight",
1729
+ "shape": [
1730
+ 2048,
1731
+ 1536
1732
+ ],
1733
+ "dtype": "float16",
1734
+ "format": "f32-to-bf16",
1735
+ "nbytes": 6291456,
1736
+ "byteOffset": 22040576
1737
+ },
1738
+ {
1739
+ "name": "model.layers.24.self_attn.o_proj.weight",
1740
+ "shape": [
1741
+ 1536,
1742
+ 1536
1743
+ ],
1744
+ "dtype": "float16",
1745
+ "format": "f32-to-bf16",
1746
+ "nbytes": 4718592,
1747
+ "byteOffset": 28332032
1748
+ },
1749
+ {
1750
+ "name": "model.layers.25.input_layernorm.weight",
1751
+ "shape": [
1752
+ 1536
1753
+ ],
1754
+ "dtype": "float16",
1755
+ "format": "f32-to-bf16",
1756
+ "nbytes": 3072,
1757
+ "byteOffset": 33050624
1758
+ },
1759
+ {
1760
+ "name": "model.layers.25.post_attention_layernorm.weight",
1761
+ "shape": [
1762
+ 1536
1763
+ ],
1764
+ "dtype": "float16",
1765
+ "format": "f32-to-bf16",
1766
+ "nbytes": 3072,
1767
+ "byteOffset": 33053696
1768
+ },
1769
+ {
1770
+ "name": "model.layers.25.self_attn.c_attn.bias",
1771
+ "shape": [
1772
+ 2048
1773
+ ],
1774
+ "dtype": "float16",
1775
+ "format": "f32-to-bf16",
1776
+ "nbytes": 4096,
1777
+ "byteOffset": 33056768
1778
+ }
1779
+ ],
1780
+ "md5sum": "dc33429dc1878212438734493b4de2aa"
1781
+ },
1782
+ {
1783
+ "dataPath": "params_shard_46.bin",
1784
+ "format": "raw-shard",
1785
+ "nbytes": 27525120,
1786
+ "records": [
1787
+ {
1788
+ "name": "model.layers.26.mlp.down_proj.weight",
1789
+ "shape": [
1790
+ 1536,
1791
+ 8960
1792
+ ],
1793
+ "dtype": "float16",
1794
+ "format": "f32-to-bf16",
1795
+ "nbytes": 27525120,
1796
+ "byteOffset": 0
1797
+ }
1798
+ ],
1799
+ "md5sum": "a5fed699a77811991fb069ec33ec8b2f"
1800
+ },
1801
+ {
1802
+ "dataPath": "params_shard_47.bin",
1803
+ "format": "raw-shard",
1804
+ "nbytes": 55050240,
1805
+ "records": [
1806
+ {
1807
+ "name": "model.layers.26.mlp.gate_up_proj.weight",
1808
+ "shape": [
1809
+ 17920,
1810
+ 1536
1811
+ ],
1812
+ "dtype": "float16",
1813
+ "format": "f32-to-bf16",
1814
+ "nbytes": 55050240,
1815
+ "byteOffset": 0
1816
+ }
1817
+ ],
1818
+ "md5sum": "d43f6f25d946d9dab2d69d525c0fca55"
1819
+ },
1820
+ {
1821
+ "dataPath": "params_shard_48.bin",
1822
+ "format": "raw-shard",
1823
+ "nbytes": 27525120,
1824
+ "records": [
1825
+ {
1826
+ "name": "model.layers.27.mlp.down_proj.weight",
1827
+ "shape": [
1828
+ 1536,
1829
+ 8960
1830
+ ],
1831
+ "dtype": "float16",
1832
+ "format": "f32-to-bf16",
1833
+ "nbytes": 27525120,
1834
+ "byteOffset": 0
1835
+ }
1836
+ ],
1837
+ "md5sum": "9104c976c3f0f79fb5ea002847b51a6a"
1838
+ },
1839
+ {
1840
+ "dataPath": "params_shard_49.bin",
1841
+ "format": "raw-shard",
1842
+ "nbytes": 55050240,
1843
+ "records": [
1844
+ {
1845
+ "name": "model.layers.27.mlp.gate_up_proj.weight",
1846
+ "shape": [
1847
+ 17920,
1848
+ 1536
1849
+ ],
1850
+ "dtype": "float16",
1851
+ "format": "f32-to-bf16",
1852
+ "nbytes": 55050240,
1853
+ "byteOffset": 0
1854
+ }
1855
+ ],
1856
+ "md5sum": "c1d597f0ccd26539aa70a0cd7b60981c"
1857
+ },
1858
+ {
1859
+ "dataPath": "params_shard_50.bin",
1860
+ "format": "raw-shard",
1861
+ "nbytes": 27525120,
1862
+ "records": [
1863
+ {
1864
+ "name": "model.layers.3.mlp.down_proj.weight",
1865
+ "shape": [
1866
+ 1536,
1867
+ 8960
1868
+ ],
1869
+ "dtype": "float16",
1870
+ "format": "f32-to-bf16",
1871
+ "nbytes": 27525120,
1872
+ "byteOffset": 0
1873
+ }
1874
+ ],
1875
+ "md5sum": "010fdd102a51b95cd310a15b471487d3"
1876
+ },
1877
+ {
1878
+ "dataPath": "params_shard_51.bin",
1879
+ "format": "raw-shard",
1880
+ "nbytes": 55050240,
1881
+ "records": [
1882
+ {
1883
+ "name": "model.layers.3.mlp.gate_up_proj.weight",
1884
+ "shape": [
1885
+ 17920,
1886
+ 1536
1887
+ ],
1888
+ "dtype": "float16",
1889
+ "format": "f32-to-bf16",
1890
+ "nbytes": 55050240,
1891
+ "byteOffset": 0
1892
+ }
1893
+ ],
1894
+ "md5sum": "598d0dc39259f1ab7a9e7ace7b47f46a"
1895
+ },
1896
+ {
1897
+ "dataPath": "params_shard_52.bin",
1898
+ "format": "raw-shard",
1899
+ "nbytes": 33060864,
1900
+ "records": [
1901
+ {
1902
+ "name": "model.layers.25.self_attn.c_attn.weight",
1903
+ "shape": [
1904
+ 2048,
1905
+ 1536
1906
+ ],
1907
+ "dtype": "float16",
1908
+ "format": "f32-to-bf16",
1909
+ "nbytes": 6291456,
1910
+ "byteOffset": 0
1911
+ },
1912
+ {
1913
+ "name": "model.layers.25.self_attn.o_proj.weight",
1914
+ "shape": [
1915
+ 1536,
1916
+ 1536
1917
+ ],
1918
+ "dtype": "float16",
1919
+ "format": "f32-to-bf16",
1920
+ "nbytes": 4718592,
1921
+ "byteOffset": 6291456
1922
+ },
1923
+ {
1924
+ "name": "model.layers.26.input_layernorm.weight",
1925
+ "shape": [
1926
+ 1536
1927
+ ],
1928
+ "dtype": "float16",
1929
+ "format": "f32-to-bf16",
1930
+ "nbytes": 3072,
1931
+ "byteOffset": 11010048
1932
+ },
1933
+ {
1934
+ "name": "model.layers.26.post_attention_layernorm.weight",
1935
+ "shape": [
1936
+ 1536
1937
+ ],
1938
+ "dtype": "float16",
1939
+ "format": "f32-to-bf16",
1940
+ "nbytes": 3072,
1941
+ "byteOffset": 11013120
1942
+ },
1943
+ {
1944
+ "name": "model.layers.26.self_attn.c_attn.bias",
1945
+ "shape": [
1946
+ 2048
1947
+ ],
1948
+ "dtype": "float16",
1949
+ "format": "f32-to-bf16",
1950
+ "nbytes": 4096,
1951
+ "byteOffset": 11016192
1952
+ },
1953
+ {
1954
+ "name": "model.layers.26.self_attn.c_attn.weight",
1955
+ "shape": [
1956
+ 2048,
1957
+ 1536
1958
+ ],
1959
+ "dtype": "float16",
1960
+ "format": "f32-to-bf16",
1961
+ "nbytes": 6291456,
1962
+ "byteOffset": 11020288
1963
+ },
1964
+ {
1965
+ "name": "model.layers.26.self_attn.o_proj.weight",
1966
+ "shape": [
1967
+ 1536,
1968
+ 1536
1969
+ ],
1970
+ "dtype": "float16",
1971
+ "format": "f32-to-bf16",
1972
+ "nbytes": 4718592,
1973
+ "byteOffset": 17311744
1974
+ },
1975
+ {
1976
+ "name": "model.layers.27.input_layernorm.weight",
1977
+ "shape": [
1978
+ 1536
1979
+ ],
1980
+ "dtype": "float16",
1981
+ "format": "f32-to-bf16",
1982
+ "nbytes": 3072,
1983
+ "byteOffset": 22030336
1984
+ },
1985
+ {
1986
+ "name": "model.layers.27.post_attention_layernorm.weight",
1987
+ "shape": [
1988
+ 1536
1989
+ ],
1990
+ "dtype": "float16",
1991
+ "format": "f32-to-bf16",
1992
+ "nbytes": 3072,
1993
+ "byteOffset": 22033408
1994
+ },
1995
+ {
1996
+ "name": "model.layers.27.self_attn.c_attn.bias",
1997
+ "shape": [
1998
+ 2048
1999
+ ],
2000
+ "dtype": "float16",
2001
+ "format": "f32-to-bf16",
2002
+ "nbytes": 4096,
2003
+ "byteOffset": 22036480
2004
+ },
2005
+ {
2006
+ "name": "model.layers.27.self_attn.c_attn.weight",
2007
+ "shape": [
2008
+ 2048,
2009
+ 1536
2010
+ ],
2011
+ "dtype": "float16",
2012
+ "format": "f32-to-bf16",
2013
+ "nbytes": 6291456,
2014
+ "byteOffset": 22040576
2015
+ },
2016
+ {
2017
+ "name": "model.layers.27.self_attn.o_proj.weight",
2018
+ "shape": [
2019
+ 1536,
2020
+ 1536
2021
+ ],
2022
+ "dtype": "float16",
2023
+ "format": "f32-to-bf16",
2024
+ "nbytes": 4718592,
2025
+ "byteOffset": 28332032
2026
+ },
2027
+ {
2028
+ "name": "model.layers.3.input_layernorm.weight",
2029
+ "shape": [
2030
+ 1536
2031
+ ],
2032
+ "dtype": "float16",
2033
+ "format": "f32-to-bf16",
2034
+ "nbytes": 3072,
2035
+ "byteOffset": 33050624
2036
+ },
2037
+ {
2038
+ "name": "model.layers.3.post_attention_layernorm.weight",
2039
+ "shape": [
2040
+ 1536
2041
+ ],
2042
+ "dtype": "float16",
2043
+ "format": "f32-to-bf16",
2044
+ "nbytes": 3072,
2045
+ "byteOffset": 33053696
2046
+ },
2047
+ {
2048
+ "name": "model.layers.3.self_attn.c_attn.bias",
2049
+ "shape": [
2050
+ 2048
2051
+ ],
2052
+ "dtype": "float16",
2053
+ "format": "f32-to-bf16",
2054
+ "nbytes": 4096,
2055
+ "byteOffset": 33056768
2056
+ }
2057
+ ],
2058
+ "md5sum": "f09bfdabd3e4dccc0ec537d453bd3ff0"
2059
+ },
2060
+ {
2061
+ "dataPath": "params_shard_53.bin",
2062
+ "format": "raw-shard",
2063
+ "nbytes": 27525120,
2064
+ "records": [
2065
+ {
2066
+ "name": "model.layers.4.mlp.down_proj.weight",
2067
+ "shape": [
2068
+ 1536,
2069
+ 8960
2070
+ ],
2071
+ "dtype": "float16",
2072
+ "format": "f32-to-bf16",
2073
+ "nbytes": 27525120,
2074
+ "byteOffset": 0
2075
+ }
2076
+ ],
2077
+ "md5sum": "1cf8b99bb81904f0fbe87954c7c090d4"
2078
+ },
2079
+ {
2080
+ "dataPath": "params_shard_54.bin",
2081
+ "format": "raw-shard",
2082
+ "nbytes": 55050240,
2083
+ "records": [
2084
+ {
2085
+ "name": "model.layers.4.mlp.gate_up_proj.weight",
2086
+ "shape": [
2087
+ 17920,
2088
+ 1536
2089
+ ],
2090
+ "dtype": "float16",
2091
+ "format": "f32-to-bf16",
2092
+ "nbytes": 55050240,
2093
+ "byteOffset": 0
2094
+ }
2095
+ ],
2096
+ "md5sum": "62d916c6fc81913257cbd4890b0c400e"
2097
+ },
2098
+ {
2099
+ "dataPath": "params_shard_55.bin",
2100
+ "format": "raw-shard",
2101
+ "nbytes": 27525120,
2102
+ "records": [
2103
+ {
2104
+ "name": "model.layers.5.mlp.down_proj.weight",
2105
+ "shape": [
2106
+ 1536,
2107
+ 8960
2108
+ ],
2109
+ "dtype": "float16",
2110
+ "format": "f32-to-bf16",
2111
+ "nbytes": 27525120,
2112
+ "byteOffset": 0
2113
+ }
2114
+ ],
2115
+ "md5sum": "a8065be5ff3a1b84fc233fa22ee79387"
2116
+ },
2117
+ {
2118
+ "dataPath": "params_shard_56.bin",
2119
+ "format": "raw-shard",
2120
+ "nbytes": 55050240,
2121
+ "records": [
2122
+ {
2123
+ "name": "model.layers.5.mlp.gate_up_proj.weight",
2124
+ "shape": [
2125
+ 17920,
2126
+ 1536
2127
+ ],
2128
+ "dtype": "float16",
2129
+ "format": "f32-to-bf16",
2130
+ "nbytes": 55050240,
2131
+ "byteOffset": 0
2132
+ }
2133
+ ],
2134
+ "md5sum": "afc16a9ee1041f8d40f4b19657550b20"
2135
+ },
2136
+ {
2137
+ "dataPath": "params_shard_57.bin",
2138
+ "format": "raw-shard",
2139
+ "nbytes": 27525120,
2140
+ "records": [
2141
+ {
2142
+ "name": "model.layers.6.mlp.down_proj.weight",
2143
+ "shape": [
2144
+ 1536,
2145
+ 8960
2146
+ ],
2147
+ "dtype": "float16",
2148
+ "format": "f32-to-bf16",
2149
+ "nbytes": 27525120,
2150
+ "byteOffset": 0
2151
+ }
2152
+ ],
2153
+ "md5sum": "50356e8b1fa958c57c3d728a4955dfa1"
2154
+ },
2155
+ {
2156
+ "dataPath": "params_shard_58.bin",
2157
+ "format": "raw-shard",
2158
+ "nbytes": 55050240,
2159
+ "records": [
2160
+ {
2161
+ "name": "model.layers.6.mlp.gate_up_proj.weight",
2162
+ "shape": [
2163
+ 17920,
2164
+ 1536
2165
+ ],
2166
+ "dtype": "float16",
2167
+ "format": "f32-to-bf16",
2168
+ "nbytes": 55050240,
2169
+ "byteOffset": 0
2170
+ }
2171
+ ],
2172
+ "md5sum": "a667254802236f0f3a49a7659f14e7c9"
2173
+ },
2174
+ {
2175
+ "dataPath": "params_shard_59.bin",
2176
+ "format": "raw-shard",
2177
+ "nbytes": 33060864,
2178
+ "records": [
2179
+ {
2180
+ "name": "model.layers.3.self_attn.c_attn.weight",
2181
+ "shape": [
2182
+ 2048,
2183
+ 1536
2184
+ ],
2185
+ "dtype": "float16",
2186
+ "format": "f32-to-bf16",
2187
+ "nbytes": 6291456,
2188
+ "byteOffset": 0
2189
+ },
2190
+ {
2191
+ "name": "model.layers.3.self_attn.o_proj.weight",
2192
+ "shape": [
2193
+ 1536,
2194
+ 1536
2195
+ ],
2196
+ "dtype": "float16",
2197
+ "format": "f32-to-bf16",
2198
+ "nbytes": 4718592,
2199
+ "byteOffset": 6291456
2200
+ },
2201
+ {
2202
+ "name": "model.layers.4.input_layernorm.weight",
2203
+ "shape": [
2204
+ 1536
2205
+ ],
2206
+ "dtype": "float16",
2207
+ "format": "f32-to-bf16",
2208
+ "nbytes": 3072,
2209
+ "byteOffset": 11010048
2210
+ },
2211
+ {
2212
+ "name": "model.layers.4.post_attention_layernorm.weight",
2213
+ "shape": [
2214
+ 1536
2215
+ ],
2216
+ "dtype": "float16",
2217
+ "format": "f32-to-bf16",
2218
+ "nbytes": 3072,
2219
+ "byteOffset": 11013120
2220
+ },
2221
+ {
2222
+ "name": "model.layers.4.self_attn.c_attn.bias",
2223
+ "shape": [
2224
+ 2048
2225
+ ],
2226
+ "dtype": "float16",
2227
+ "format": "f32-to-bf16",
2228
+ "nbytes": 4096,
2229
+ "byteOffset": 11016192
2230
+ },
2231
+ {
2232
+ "name": "model.layers.4.self_attn.c_attn.weight",
2233
+ "shape": [
2234
+ 2048,
2235
+ 1536
2236
+ ],
2237
+ "dtype": "float16",
2238
+ "format": "f32-to-bf16",
2239
+ "nbytes": 6291456,
2240
+ "byteOffset": 11020288
2241
+ },
2242
+ {
2243
+ "name": "model.layers.4.self_attn.o_proj.weight",
2244
+ "shape": [
2245
+ 1536,
2246
+ 1536
2247
+ ],
2248
+ "dtype": "float16",
2249
+ "format": "f32-to-bf16",
2250
+ "nbytes": 4718592,
2251
+ "byteOffset": 17311744
2252
+ },
2253
+ {
2254
+ "name": "model.layers.5.input_layernorm.weight",
2255
+ "shape": [
2256
+ 1536
2257
+ ],
2258
+ "dtype": "float16",
2259
+ "format": "f32-to-bf16",
2260
+ "nbytes": 3072,
2261
+ "byteOffset": 22030336
2262
+ },
2263
+ {
2264
+ "name": "model.layers.5.post_attention_layernorm.weight",
2265
+ "shape": [
2266
+ 1536
2267
+ ],
2268
+ "dtype": "float16",
2269
+ "format": "f32-to-bf16",
2270
+ "nbytes": 3072,
2271
+ "byteOffset": 22033408
2272
+ },
2273
+ {
2274
+ "name": "model.layers.5.self_attn.c_attn.bias",
2275
+ "shape": [
2276
+ 2048
2277
+ ],
2278
+ "dtype": "float16",
2279
+ "format": "f32-to-bf16",
2280
+ "nbytes": 4096,
2281
+ "byteOffset": 22036480
2282
+ },
2283
+ {
2284
+ "name": "model.layers.5.self_attn.c_attn.weight",
2285
+ "shape": [
2286
+ 2048,
2287
+ 1536
2288
+ ],
2289
+ "dtype": "float16",
2290
+ "format": "f32-to-bf16",
2291
+ "nbytes": 6291456,
2292
+ "byteOffset": 22040576
2293
+ },
2294
+ {
2295
+ "name": "model.layers.5.self_attn.o_proj.weight",
2296
+ "shape": [
2297
+ 1536,
2298
+ 1536
2299
+ ],
2300
+ "dtype": "float16",
2301
+ "format": "f32-to-bf16",
2302
+ "nbytes": 4718592,
2303
+ "byteOffset": 28332032
2304
+ },
2305
+ {
2306
+ "name": "model.layers.6.input_layernorm.weight",
2307
+ "shape": [
2308
+ 1536
2309
+ ],
2310
+ "dtype": "float16",
2311
+ "format": "f32-to-bf16",
2312
+ "nbytes": 3072,
2313
+ "byteOffset": 33050624
2314
+ },
2315
+ {
2316
+ "name": "model.layers.6.post_attention_layernorm.weight",
2317
+ "shape": [
2318
+ 1536
2319
+ ],
2320
+ "dtype": "float16",
2321
+ "format": "f32-to-bf16",
2322
+ "nbytes": 3072,
2323
+ "byteOffset": 33053696
2324
+ },
2325
+ {
2326
+ "name": "model.layers.6.self_attn.c_attn.bias",
2327
+ "shape": [
2328
+ 2048
2329
+ ],
2330
+ "dtype": "float16",
2331
+ "format": "f32-to-bf16",
2332
+ "nbytes": 4096,
2333
+ "byteOffset": 33056768
2334
+ }
2335
+ ],
2336
+ "md5sum": "d956f7171d618981966fc174cae6f61d"
2337
+ },
2338
+ {
2339
+ "dataPath": "params_shard_60.bin",
2340
+ "format": "raw-shard",
2341
+ "nbytes": 27525120,
2342
+ "records": [
2343
+ {
2344
+ "name": "model.layers.7.mlp.down_proj.weight",
2345
+ "shape": [
2346
+ 1536,
2347
+ 8960
2348
+ ],
2349
+ "dtype": "float16",
2350
+ "format": "f32-to-bf16",
2351
+ "nbytes": 27525120,
2352
+ "byteOffset": 0
2353
+ }
2354
+ ],
2355
+ "md5sum": "645216271ed732eee9146ef4842f45a7"
2356
+ },
2357
+ {
2358
+ "dataPath": "params_shard_61.bin",
2359
+ "format": "raw-shard",
2360
+ "nbytes": 55050240,
2361
+ "records": [
2362
+ {
2363
+ "name": "model.layers.7.mlp.gate_up_proj.weight",
2364
+ "shape": [
2365
+ 17920,
2366
+ 1536
2367
+ ],
2368
+ "dtype": "float16",
2369
+ "format": "f32-to-bf16",
2370
+ "nbytes": 55050240,
2371
+ "byteOffset": 0
2372
+ }
2373
+ ],
2374
+ "md5sum": "ded72450ae4b608394f76c94faa77944"
2375
+ },
2376
+ {
2377
+ "dataPath": "params_shard_62.bin",
2378
+ "format": "raw-shard",
2379
+ "nbytes": 27525120,
2380
+ "records": [
2381
+ {
2382
+ "name": "model.layers.8.mlp.down_proj.weight",
2383
+ "shape": [
2384
+ 1536,
2385
+ 8960
2386
+ ],
2387
+ "dtype": "float16",
2388
+ "format": "f32-to-bf16",
2389
+ "nbytes": 27525120,
2390
+ "byteOffset": 0
2391
+ }
2392
+ ],
2393
+ "md5sum": "b4c989a18b750af37b85a4ad69ffefbd"
2394
+ },
2395
+ {
2396
+ "dataPath": "params_shard_63.bin",
2397
+ "format": "raw-shard",
2398
+ "nbytes": 55050240,
2399
+ "records": [
2400
+ {
2401
+ "name": "model.layers.8.mlp.gate_up_proj.weight",
2402
+ "shape": [
2403
+ 17920,
2404
+ 1536
2405
+ ],
2406
+ "dtype": "float16",
2407
+ "format": "f32-to-bf16",
2408
+ "nbytes": 55050240,
2409
+ "byteOffset": 0
2410
+ }
2411
+ ],
2412
+ "md5sum": "9dbee59512c02d9c2207ca383daa2d6d"
2413
+ },
2414
+ {
2415
+ "dataPath": "params_shard_64.bin",
2416
+ "format": "raw-shard",
2417
+ "nbytes": 27525120,
2418
+ "records": [
2419
+ {
2420
+ "name": "model.layers.9.mlp.down_proj.weight",
2421
+ "shape": [
2422
+ 1536,
2423
+ 8960
2424
+ ],
2425
+ "dtype": "float16",
2426
+ "format": "f32-to-bf16",
2427
+ "nbytes": 27525120,
2428
+ "byteOffset": 0
2429
+ }
2430
+ ],
2431
+ "md5sum": "1b4766e9f4a576f525569989ea262855"
2432
+ },
2433
+ {
2434
+ "dataPath": "params_shard_65.bin",
2435
+ "format": "raw-shard",
2436
+ "nbytes": 55050240,
2437
+ "records": [
2438
+ {
2439
+ "name": "model.layers.9.mlp.gate_up_proj.weight",
2440
+ "shape": [
2441
+ 17920,
2442
+ 1536
2443
+ ],
2444
+ "dtype": "float16",
2445
+ "format": "f32-to-bf16",
2446
+ "nbytes": 55050240,
2447
+ "byteOffset": 0
2448
+ }
2449
+ ],
2450
+ "md5sum": "d9828ef62b87d02af740526feadb67e7"
2451
+ },
2452
+ {
2453
+ "dataPath": "params_shard_66.bin",
2454
+ "format": "raw-shard",
2455
+ "nbytes": 33060864,
2456
+ "records": [
2457
+ {
2458
+ "name": "model.layers.6.self_attn.c_attn.weight",
2459
+ "shape": [
2460
+ 2048,
2461
+ 1536
2462
+ ],
2463
+ "dtype": "float16",
2464
+ "format": "f32-to-bf16",
2465
+ "nbytes": 6291456,
2466
+ "byteOffset": 0
2467
+ },
2468
+ {
2469
+ "name": "model.layers.6.self_attn.o_proj.weight",
2470
+ "shape": [
2471
+ 1536,
2472
+ 1536
2473
+ ],
2474
+ "dtype": "float16",
2475
+ "format": "f32-to-bf16",
2476
+ "nbytes": 4718592,
2477
+ "byteOffset": 6291456
2478
+ },
2479
+ {
2480
+ "name": "model.layers.7.input_layernorm.weight",
2481
+ "shape": [
2482
+ 1536
2483
+ ],
2484
+ "dtype": "float16",
2485
+ "format": "f32-to-bf16",
2486
+ "nbytes": 3072,
2487
+ "byteOffset": 11010048
2488
+ },
2489
+ {
2490
+ "name": "model.layers.7.post_attention_layernorm.weight",
2491
+ "shape": [
2492
+ 1536
2493
+ ],
2494
+ "dtype": "float16",
2495
+ "format": "f32-to-bf16",
2496
+ "nbytes": 3072,
2497
+ "byteOffset": 11013120
2498
+ },
2499
+ {
2500
+ "name": "model.layers.7.self_attn.c_attn.bias",
2501
+ "shape": [
2502
+ 2048
2503
+ ],
2504
+ "dtype": "float16",
2505
+ "format": "f32-to-bf16",
2506
+ "nbytes": 4096,
2507
+ "byteOffset": 11016192
2508
+ },
2509
+ {
2510
+ "name": "model.layers.7.self_attn.c_attn.weight",
2511
+ "shape": [
2512
+ 2048,
2513
+ 1536
2514
+ ],
2515
+ "dtype": "float16",
2516
+ "format": "f32-to-bf16",
2517
+ "nbytes": 6291456,
2518
+ "byteOffset": 11020288
2519
+ },
2520
+ {
2521
+ "name": "model.layers.7.self_attn.o_proj.weight",
2522
+ "shape": [
2523
+ 1536,
2524
+ 1536
2525
+ ],
2526
+ "dtype": "float16",
2527
+ "format": "f32-to-bf16",
2528
+ "nbytes": 4718592,
2529
+ "byteOffset": 17311744
2530
+ },
2531
+ {
2532
+ "name": "model.layers.8.input_layernorm.weight",
2533
+ "shape": [
2534
+ 1536
2535
+ ],
2536
+ "dtype": "float16",
2537
+ "format": "f32-to-bf16",
2538
+ "nbytes": 3072,
2539
+ "byteOffset": 22030336
2540
+ },
2541
+ {
2542
+ "name": "model.layers.8.post_attention_layernorm.weight",
2543
+ "shape": [
2544
+ 1536
2545
+ ],
2546
+ "dtype": "float16",
2547
+ "format": "f32-to-bf16",
2548
+ "nbytes": 3072,
2549
+ "byteOffset": 22033408
2550
+ },
2551
+ {
2552
+ "name": "model.layers.8.self_attn.c_attn.bias",
2553
+ "shape": [
2554
+ 2048
2555
+ ],
2556
+ "dtype": "float16",
2557
+ "format": "f32-to-bf16",
2558
+ "nbytes": 4096,
2559
+ "byteOffset": 22036480
2560
+ },
2561
+ {
2562
+ "name": "model.layers.8.self_attn.c_attn.weight",
2563
+ "shape": [
2564
+ 2048,
2565
+ 1536
2566
+ ],
2567
+ "dtype": "float16",
2568
+ "format": "f32-to-bf16",
2569
+ "nbytes": 6291456,
2570
+ "byteOffset": 22040576
2571
+ },
2572
+ {
2573
+ "name": "model.layers.8.self_attn.o_proj.weight",
2574
+ "shape": [
2575
+ 1536,
2576
+ 1536
2577
+ ],
2578
+ "dtype": "float16",
2579
+ "format": "f32-to-bf16",
2580
+ "nbytes": 4718592,
2581
+ "byteOffset": 28332032
2582
+ },
2583
+ {
2584
+ "name": "model.layers.9.input_layernorm.weight",
2585
+ "shape": [
2586
+ 1536
2587
+ ],
2588
+ "dtype": "float16",
2589
+ "format": "f32-to-bf16",
2590
+ "nbytes": 3072,
2591
+ "byteOffset": 33050624
2592
+ },
2593
+ {
2594
+ "name": "model.layers.9.post_attention_layernorm.weight",
2595
+ "shape": [
2596
+ 1536
2597
+ ],
2598
+ "dtype": "float16",
2599
+ "format": "f32-to-bf16",
2600
+ "nbytes": 3072,
2601
+ "byteOffset": 33053696
2602
+ },
2603
+ {
2604
+ "name": "model.layers.9.self_attn.c_attn.bias",
2605
+ "shape": [
2606
+ 2048
2607
+ ],
2608
+ "dtype": "float16",
2609
+ "format": "f32-to-bf16",
2610
+ "nbytes": 4096,
2611
+ "byteOffset": 33056768
2612
+ }
2613
+ ],
2614
+ "md5sum": "ba69a1abb6f090c015317fea9b3a3408"
2615
+ },
2616
+ {
2617
+ "dataPath": "params_shard_67.bin",
2618
+ "format": "raw-shard",
2619
+ "nbytes": 11013120,
2620
+ "records": [
2621
+ {
2622
+ "name": "model.layers.9.self_attn.c_attn.weight",
2623
+ "shape": [
2624
+ 2048,
2625
+ 1536
2626
+ ],
2627
+ "dtype": "float16",
2628
+ "format": "f32-to-bf16",
2629
+ "nbytes": 6291456,
2630
+ "byteOffset": 0
2631
+ },
2632
+ {
2633
+ "name": "model.layers.9.self_attn.o_proj.weight",
2634
+ "shape": [
2635
+ 1536,
2636
+ 1536
2637
+ ],
2638
+ "dtype": "float16",
2639
+ "format": "f32-to-bf16",
2640
+ "nbytes": 4718592,
2641
+ "byteOffset": 6291456
2642
+ },
2643
+ {
2644
+ "name": "model.norm.weight",
2645
+ "shape": [
2646
+ 1536
2647
+ ],
2648
+ "dtype": "float16",
2649
+ "format": "f32-to-bf16",
2650
+ "nbytes": 3072,
2651
+ "byteOffset": 11010048
2652
+ }
2653
+ ],
2654
+ "md5sum": "9b206085ecacfac239a6d4045981af0b"
2655
+ }
2656
+ ]
2657
+ }
params_shard_0.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cca68befcc8201afc0eb54623dd20bd2af92acfe3cff767e6f8e6c0ddad2a397
3
+ size 466747392
params_shard_1.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e55610c68685326d482c594ff3bb16141e71a0d219fe729211562ab630953c6e
3
+ size 466747392
params_shard_10.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:38cc1a8ee64126bc13c1443ed8e183660870ee13ded7263712c8cca4e5fa76f3
3
+ size 33060864
params_shard_11.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a927d17c757e520a6b6d94ddb86167bb9ff03c7bda4968aee28748994994a14
3
+ size 27525120
params_shard_12.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf9bad949697b6da8fa64a7b50e087a396983d7add65c4cb3e6061d291b0491b
3
+ size 55050240
params_shard_13.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:945549e4c0d593fecbc010eb023b7838eecded68eef0ba18767d899f0ea8b1e5
3
+ size 27525120
params_shard_14.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8fff9cd0a524f0f3a0f507c79690f432cb4345b5e1a174c62e59aaafe34c2d5f
3
+ size 55050240
params_shard_15.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:af28f09aaae89875033642a1f844be79bad53efc785f434cabfae2bb5ffdc944
3
+ size 27525120
params_shard_16.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c35caa4c62e4e6ff5331f9eddacedb87ecd2d028eac90e58bbbe6f4f34fa71ce
3
+ size 55050240
params_shard_17.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6cfbbc25a2dd881b5bf429742ece25453009e06ba534d6e37082fd44c798446a
3
+ size 33060864
params_shard_18.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aecc9aceabbb32b6e0b45deeb918a368606226a11a9489f1b1526ae931560458
3
+ size 27525120
params_shard_19.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:666678f8361aeeabdf5f117503c4c443e1ef3272c3366f6c3bc51884497ebe7a
3
+ size 55050240
params_shard_2.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:384913d24986a8e831898c03226167a4e286cfb4d8845b0f30ac1341880b9b80
3
+ size 55050240
params_shard_20.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:554f1b8cd2160c7c1dd0dc934458ae507dc9bf4249514105df4ba4985a408669
3
+ size 27525120
params_shard_21.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:909599756505820c805f0385589a7bfd9d841a38844423c665e9f4c99822bd3d
3
+ size 55050240
params_shard_22.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:79620ae2d20278c08f005341b85367b8e6362d50ceb4f9071bb376fafeab9436
3
+ size 27525120
params_shard_23.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:678b66722d61726dd654b318191f31844450f964469755c1661f13995e2d1053
3
+ size 55050240
params_shard_24.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:92eec36d7915ec2cb957f038088f7832eb315de2fe5b84df8e61b9a902a55b3b
3
+ size 33060864
params_shard_25.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc5bcd539df900d404787c84d24fec4d835cad093ddd25cbd73e012803217525
3
+ size 27525120
params_shard_26.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e9c1fd6348607f64df6653242d66beca19969d3772095d23c9735b6573be5bfe
3
+ size 55050240
params_shard_27.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:68e3771b069d3d0962b9c9aa626c325686581c568b73d1e095ff47ea13855602
3
+ size 27525120
params_shard_28.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fb45484edf21f78d2d5a4a3615fd25298d3015560d1a496bc8650dab0745eae0
3
+ size 55050240
params_shard_29.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bd0595532b11a8141dc536e0618bad907eb686d8f56401d859522c9ca1bb0dd0
3
+ size 27525120
params_shard_3.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7326c1d95d5d08d3eea489e613847992290a33bb2f0c56ac6720bfb0854db471
3
+ size 27535360
params_shard_30.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3c73dd03c57f831eb98b4d7feec58894a4708a0aac10f46ee43084628db705f6
3
+ size 55050240
params_shard_31.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d9040f278fbb6e578271fd29820456e651e83bfc381e194ebbdd14f1d1b553ab
3
+ size 33060864
params_shard_32.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30e82d6160972c9631fc4186c490237847fae0b3cf117ae221b177b89143c1a6
3
+ size 27525120
params_shard_33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:efbf6615c21c4b392085b584c62ebc0106db468991978f1f291c5ada420b6ed5
3
+ size 55050240
params_shard_34.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66955e891af65d45afdc68f90f13eaa8d8a0c97b1556e39be3a823f0f7500aa2
3
+ size 27525120
params_shard_35.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81d17c7b976ac2c77cecd8fb29474a093017c94020b8de8a65c7161d4790e762
3
+ size 55050240
params_shard_36.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9f8ccd71e2523151926e5f8eb489d03b0ac4febabc7f93f7d0a8fae1128c6ca3
3
+ size 27525120
params_shard_37.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bdedaa8e70b021a9abad90569a6ede60dc179e4c9e2c36bd6de74ca8e04ff76b
3
+ size 55050240
params_shard_38.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce31d48c5cf8e38cbb2f4d6a38b6d4741ad0b2d44433de43f8ddbbbb6b15c5b3
3
+ size 33060864
params_shard_39.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89a3e6a1ce529b46d4b2addcdf95e1b0d8ee29fe482fa1fb60b83dfcf6d06429
3
+ size 27525120
params_shard_4.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9b20777c2c303b7cdc4f763d85424eb874a0b68e0905c3777544b15048903db1
3
+ size 27525120
params_shard_40.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:69e619c5a29baaa1ff841413d3d90d69dca43ad25baddb00c4c9cad99f5c9e61
3
+ size 55050240
params_shard_41.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e974a9a6ecc970e01d61c8eb28076c6f49433b905896e6bc8bb3988e3f9de0cc
3
+ size 27525120
params_shard_42.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:762097e49845c377e59d424b4bbb1db31f6b4e252ed3fda9722c491e01cd7f69
3
+ size 55050240
params_shard_43.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a49250d78e90a5aa34e31229a9d28652fb8df2fa7e6cb45e326902a85b9c8b7
3
+ size 27525120
params_shard_44.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:43cc8326c30fb63c68605f818f24472bafeebbade538d7904e553887d527c348
3
+ size 55050240
params_shard_45.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bab893d95cd3527555a4148f7f47961c3b8015ffbdb0ef4e061811b5400dd966
3
+ size 33060864
params_shard_46.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:27465df3716efbaed682c2f74ac1e7c3f75e117a8133ab2ad1273412f8d9f2a7
3
+ size 27525120
params_shard_47.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dcbdad0412505b00f20063443f0e25ca711d5c034c61ac429147c2bc3a797969
3
+ size 55050240
params_shard_48.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b2443d1edc706fc1bcc81857cace7537b5ff62757ad088bf9d402166d8df49ad
3
+ size 27525120
params_shard_49.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:31e20866bf6509c5c9b0e984e12c082421645137b2a91bc98b1d9df55775bcfa
3
+ size 55050240
params_shard_5.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:00ab864e080695fb64032120ff87f3e0e40e38979216a8489e47bc0b54930e62
3
+ size 55050240
params_shard_50.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e62ab85543dd473f94e95ee4ce9ece1a6fcb352d48190c2b3f9224650a6152e
3
+ size 27525120
params_shard_51.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb2a0d981729068d0d2bed59d09e7c5342130a14118fd8bc6bc340df029bbf69
3
+ size 55050240