Hugging Face's logo

lizhongzhi2022
/
TLDR_code_release

Safetensors
qwen2
Model card Files Community
lizhongzhi2022 commited on
Commit
c70e138
·
verified ·
1 Parent(s): 9582862

Upload folder using huggingface_hub

Browse files
Files changed (28) hide show
  1. .gitattributes +1 -0
  2. adapter_model/config.json +31 -0
  3. adapter_model/generation_config.json +9 -0
  4. adapter_model/model.safetensors +3 -0
  5. config.json +31 -0
  6. generation_config.json +9 -0
  7. latest +1 -0
  8. long2short_proportions.json +365 -0
  9. model-00001-of-00004.safetensors +3 -0
  10. model-00002-of-00004.safetensors +3 -0
  11. model-00003-of-00004.safetensors +3 -0
  12. model-00004-of-00004.safetensors +3 -0
  13. model.safetensors.index.json +346 -0
  14. rng_state_0.pth +3 -0
  15. rng_state_1.pth +3 -0
  16. rng_state_2.pth +3 -0
  17. rng_state_3.pth +3 -0
  18. rng_state_4.pth +3 -0
  19. rng_state_5.pth +3 -0
  20. rng_state_6.pth +3 -0
  21. rng_state_7.pth +3 -0
  22. scheduler.pt +3 -0
  23. special_tokens_map.json +23 -0
  24. tokenizer.json +3 -0
  25. tokenizer_config.json +195 -0
  26. trainer_state.json +2113 -0
  27. training_args.bin +3 -0
  28. zero_to_fp32.py +674 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
adapter_model/config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/cpfs/user/lizhongzhi/huggingface_model/huggingface_model/DeepSeek-R1-Distill-Qwen-7B/",
3
+ "architectures": [
4
+ "Qwen2ForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "eos_token_id": 151643,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 3584,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 18944,
13
+ "max_position_embeddings": 131072,
14
+ "max_window_layers": 28,
15
+ "model_type": "qwen2",
16
+ "num_attention_heads": 28,
17
+ "num_hidden_layers": 28,
18
+ "num_key_value_heads": 4,
19
+ "pad_token_id": 151643,
20
+ "rms_norm_eps": 1e-06,
21
+ "rope_scaling": null,
22
+ "rope_theta": 10000,
23
+ "sliding_window": null,
24
+ "tie_word_embeddings": false,
25
+ "torch_dtype": "bfloat16",
26
+ "transformers_version": "4.46.3",
27
+ "use_cache": true,
28
+ "use_mrope": false,
29
+ "use_sliding_window": false,
30
+ "vocab_size": 152064
31
+ }
adapter_model/generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 151646,
4
+ "do_sample": true,
5
+ "eos_token_id": 151643,
6
+ "temperature": 0.6,
7
+ "top_p": 0.95,
8
+ "transformers_version": "4.46.3"
9
+ }
adapter_model/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c395ebed7ddc70201e0286c0c4e97807244814db18d40bf2bd3b6dd88c08f7b
3
+ size 701024
config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/cpfs/user/lizhongzhi/huggingface_model/huggingface_model/DeepSeek-R1-Distill-Qwen-7B/",
3
+ "architectures": [
4
+ "Qwen2ForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "eos_token_id": 151643,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 3584,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 18944,
13
+ "max_position_embeddings": 131072,
14
+ "max_window_layers": 28,
15
+ "model_type": "qwen2",
16
+ "num_attention_heads": 28,
17
+ "num_hidden_layers": 28,
18
+ "num_key_value_heads": 4,
19
+ "pad_token_id": 151643,
20
+ "rms_norm_eps": 1e-06,
21
+ "rope_scaling": null,
22
+ "rope_theta": 10000,
23
+ "sliding_window": null,
24
+ "tie_word_embeddings": false,
25
+ "torch_dtype": "bfloat16",
26
+ "transformers_version": "4.46.3",
27
+ "use_cache": true,
28
+ "use_mrope": false,
29
+ "use_sliding_window": false,
30
+ "vocab_size": 152064
31
+ }
generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 151646,
4
+ "do_sample": true,
5
+ "eos_token_id": 151643,
6
+ "temperature": 0.6,
7
+ "top_p": 0.95,
8
+ "transformers_version": "4.46.3"
9
+ }
latest ADDED
@@ -0,0 +1 @@
 
 
1
+ global_step256
long2short_proportions.json ADDED
@@ -0,0 +1,365 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "global_step": 0,
4
+ "cot_domain_weight": [
5
+ 0.8,
6
+ 0.2
7
+ ],
8
+ "cot_domain_name": [
9
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
10
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
11
+ ]
12
+ },
13
+ {
14
+ "global_step": 8,
15
+ "cot_domain_weight": [
16
+ 0.81966233253479,
17
+ 0.18033766746520996
18
+ ],
19
+ "cot_domain_name": [
20
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
21
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
22
+ ]
23
+ },
24
+ {
25
+ "global_step": 16,
26
+ "cot_domain_weight": [
27
+ 0.7872583270072937,
28
+ 0.2127416729927063
29
+ ],
30
+ "cot_domain_name": [
31
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
32
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
33
+ ]
34
+ },
35
+ {
36
+ "global_step": 24,
37
+ "cot_domain_weight": [
38
+ 0.7460198998451233,
39
+ 0.2539801001548767
40
+ ],
41
+ "cot_domain_name": [
42
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
43
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
44
+ ]
45
+ },
46
+ {
47
+ "global_step": 32,
48
+ "cot_domain_weight": [
49
+ 0.6744258410715319,
50
+ 0.32557415892846814
51
+ ],
52
+ "cot_domain_name": [
53
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
54
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
55
+ ]
56
+ },
57
+ {
58
+ "global_step": 40,
59
+ "cot_domain_weight": [
60
+ 0.5970645546913147,
61
+ 0.4029354453086853
62
+ ],
63
+ "cot_domain_name": [
64
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
65
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
66
+ ]
67
+ },
68
+ {
69
+ "global_step": 48,
70
+ "cot_domain_weight": [
71
+ 0.3999738454834315,
72
+ 0.6000261545165685
73
+ ],
74
+ "cot_domain_name": [
75
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
76
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
77
+ ]
78
+ },
79
+ {
80
+ "global_step": 56,
81
+ "cot_domain_weight": [
82
+ 0.2720071835522165,
83
+ 0.7279928164477835
84
+ ],
85
+ "cot_domain_name": [
86
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
87
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
88
+ ]
89
+ },
90
+ {
91
+ "global_step": 64,
92
+ "cot_domain_weight": [
93
+ 0.2883644063798553,
94
+ 0.7116355936201447
95
+ ],
96
+ "cot_domain_name": [
97
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
98
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
99
+ ]
100
+ },
101
+ {
102
+ "global_step": 72,
103
+ "cot_domain_weight": [
104
+ 0.3323897124180455,
105
+ 0.6676102875819545
106
+ ],
107
+ "cot_domain_name": [
108
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
109
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
110
+ ]
111
+ },
112
+ {
113
+ "global_step": 80,
114
+ "cot_domain_weight": [
115
+ 0.3198286903057673,
116
+ 0.6801713096942327
117
+ ],
118
+ "cot_domain_name": [
119
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
120
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
121
+ ]
122
+ },
123
+ {
124
+ "global_step": 88,
125
+ "cot_domain_weight": [
126
+ 0.30956872162632476,
127
+ 0.6904312783736752
128
+ ],
129
+ "cot_domain_name": [
130
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
131
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
132
+ ]
133
+ },
134
+ {
135
+ "global_step": 96,
136
+ "cot_domain_weight": [
137
+ 0.28148205765974865,
138
+ 0.7185179423402513
139
+ ],
140
+ "cot_domain_name": [
141
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
142
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
143
+ ]
144
+ },
145
+ {
146
+ "global_step": 104,
147
+ "cot_domain_weight": [
148
+ 0.19276975382521383,
149
+ 0.8072302461747862
150
+ ],
151
+ "cot_domain_name": [
152
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
153
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
154
+ ]
155
+ },
156
+ {
157
+ "global_step": 112,
158
+ "cot_domain_weight": [
159
+ 0.11667832421803193,
160
+ 0.8833216757819681
161
+ ],
162
+ "cot_domain_name": [
163
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
164
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
165
+ ]
166
+ },
167
+ {
168
+ "global_step": 120,
169
+ "cot_domain_weight": [
170
+ 0.10653018285729127,
171
+ 0.8934698171427087
172
+ ],
173
+ "cot_domain_name": [
174
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
175
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
176
+ ]
177
+ },
178
+ {
179
+ "global_step": 128,
180
+ "cot_domain_weight": [
181
+ 0.08691881046262705,
182
+ 0.9130811895373729
183
+ ],
184
+ "cot_domain_name": [
185
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
186
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
187
+ ]
188
+ },
189
+ {
190
+ "global_step": 136,
191
+ "cot_domain_weight": [
192
+ 0.07620099413993937,
193
+ 0.9237990058600606
194
+ ],
195
+ "cot_domain_name": [
196
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
197
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
198
+ ]
199
+ },
200
+ {
201
+ "global_step": 144,
202
+ "cot_domain_weight": [
203
+ 0.09678315443384017,
204
+ 0.9032168455661598
205
+ ],
206
+ "cot_domain_name": [
207
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
208
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
209
+ ]
210
+ },
211
+ {
212
+ "global_step": 152,
213
+ "cot_domain_weight": [
214
+ 0.06039547920227051,
215
+ 0.9396045207977295
216
+ ],
217
+ "cot_domain_name": [
218
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
219
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
220
+ ]
221
+ },
222
+ {
223
+ "global_step": 160,
224
+ "cot_domain_weight": [
225
+ 0.04663034069109073,
226
+ 0.9533696593089093
227
+ ],
228
+ "cot_domain_name": [
229
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
230
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
231
+ ]
232
+ },
233
+ {
234
+ "global_step": 168,
235
+ "cot_domain_weight": [
236
+ 0.026384488927624624,
237
+ 0.9736155110723754
238
+ ],
239
+ "cot_domain_name": [
240
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
241
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
242
+ ]
243
+ },
244
+ {
245
+ "global_step": 176,
246
+ "cot_domain_weight": [
247
+ 0.01369204708991822,
248
+ 0.9863079529100818
249
+ ],
250
+ "cot_domain_name": [
251
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
252
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
253
+ ]
254
+ },
255
+ {
256
+ "global_step": 184,
257
+ "cot_domain_weight": [
258
+ 0.008808859025084685,
259
+ 0.9911911409749153
260
+ ],
261
+ "cot_domain_name": [
262
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
263
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
264
+ ]
265
+ },
266
+ {
267
+ "global_step": 192,
268
+ "cot_domain_weight": [
269
+ 0.008046488434985519,
270
+ 0.9919535115650144
271
+ ],
272
+ "cot_domain_name": [
273
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
274
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
275
+ ]
276
+ },
277
+ {
278
+ "global_step": 200,
279
+ "cot_domain_weight": [
280
+ 0.006645676632023577,
281
+ 0.9933543233679765
282
+ ],
283
+ "cot_domain_name": [
284
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
285
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
286
+ ]
287
+ },
288
+ {
289
+ "global_step": 208,
290
+ "cot_domain_weight": [
291
+ 0.006015583141017519,
292
+ 0.9939844168589825
293
+ ],
294
+ "cot_domain_name": [
295
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
296
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
297
+ ]
298
+ },
299
+ {
300
+ "global_step": 216,
301
+ "cot_domain_weight": [
302
+ 0.004511566495152915,
303
+ 0.9954884335048471
304
+ ],
305
+ "cot_domain_name": [
306
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
307
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
308
+ ]
309
+ },
310
+ {
311
+ "global_step": 224,
312
+ "cot_domain_weight": [
313
+ 0.002732270716591211,
314
+ 0.9972677292834088
315
+ ],
316
+ "cot_domain_name": [
317
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
318
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
319
+ ]
320
+ },
321
+ {
322
+ "global_step": 232,
323
+ "cot_domain_weight": [
324
+ 0.0021831512748239114,
325
+ 0.9978168487251761
326
+ ],
327
+ "cot_domain_name": [
328
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
329
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
330
+ ]
331
+ },
332
+ {
333
+ "global_step": 240,
334
+ "cot_domain_weight": [
335
+ 0.0022364268058571715,
336
+ 0.9977635731941429
337
+ ],
338
+ "cot_domain_name": [
339
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
340
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
341
+ ]
342
+ },
343
+ {
344
+ "global_step": 248,
345
+ "cot_domain_weight": [
346
+ 0.003314645357360453,
347
+ 0.9966853546426395
348
+ ],
349
+ "cot_domain_name": [
350
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
351
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
352
+ ]
353
+ },
354
+ {
355
+ "global_step": 256,
356
+ "cot_domain_weight": [
357
+ 0.0036412973637667506,
358
+ 0.9963587026362333
359
+ ],
360
+ "cot_domain_name": [
361
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/long2short_prm12k_gsm8k_merge_data/dynamic_long_short_data/gsm8k_prm12k_s1/prm12k_gsm8k_short_shortest.jsonl",
362
+ "/cpfs/user/lizhongzhi/data/data/gsm8k/liang/s1_7b_r1_prompt.jsonl"
363
+ ]
364
+ }
365
+ ]
model-00001-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd3a2c81fafdb17a408b824be2275f058cb26c4e480e3e735a0f15aa19e14151
3
+ size 4877660776
model-00002-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3dd211dd8f37c904872ea4ed8153c6bd012a34630da2a664d8ebdbaf76d4d15e
3
+ size 4932751008
model-00003-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a2faecbc251c1baceeadb56f43a7fc3f7224b771a9665ad74609069f5c9a934
3
+ size 4330865200
model-00004-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e1e94c33e6f48f80370536173c5059e771a69081642e879e48f40fe67bcb990a
3
+ size 1089994880
model.safetensors.index.json ADDED
@@ -0,0 +1,346 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 15231233024
4
+ },
5
+ "weight_map": {
6
+ "lm_head.weight": "model-00004-of-00004.safetensors",
7
+ "model.embed_tokens.weight": "model-00001-of-00004.safetensors",
8
+ "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
9
+ "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
10
+ "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
11
+ "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
12
+ "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
13
+ "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
14
+ "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
15
+ "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
16
+ "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
17
+ "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
18
+ "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
19
+ "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
20
+ "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
21
+ "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
22
+ "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
23
+ "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
24
+ "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
25
+ "model.layers.1.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
26
+ "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
27
+ "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
28
+ "model.layers.1.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
29
+ "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
30
+ "model.layers.1.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
31
+ "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
32
+ "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
33
+ "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
34
+ "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
35
+ "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
36
+ "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
37
+ "model.layers.10.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
38
+ "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
39
+ "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
40
+ "model.layers.10.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
41
+ "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
42
+ "model.layers.10.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
43
+ "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
44
+ "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
45
+ "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
46
+ "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
47
+ "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
48
+ "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
49
+ "model.layers.11.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
50
+ "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
51
+ "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
52
+ "model.layers.11.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
53
+ "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
54
+ "model.layers.11.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
55
+ "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
56
+ "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
57
+ "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
58
+ "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
59
+ "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
60
+ "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
61
+ "model.layers.12.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
62
+ "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
63
+ "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
64
+ "model.layers.12.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
65
+ "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
66
+ "model.layers.12.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
67
+ "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
68
+ "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
69
+ "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
70
+ "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
71
+ "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
72
+ "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
73
+ "model.layers.13.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
74
+ "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
75
+ "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
76
+ "model.layers.13.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
77
+ "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
78
+ "model.layers.13.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
79
+ "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
80
+ "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors",
81
+ "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
82
+ "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
83
+ "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
84
+ "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
85
+ "model.layers.14.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
86
+ "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
87
+ "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
88
+ "model.layers.14.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
89
+ "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
90
+ "model.layers.14.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
91
+ "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
92
+ "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors",
93
+ "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
94
+ "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
95
+ "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
96
+ "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
97
+ "model.layers.15.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
98
+ "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
99
+ "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
100
+ "model.layers.15.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
101
+ "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
102
+ "model.layers.15.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
103
+ "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
104
+ "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors",
105
+ "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
106
+ "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
107
+ "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
108
+ "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
109
+ "model.layers.16.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
110
+ "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
111
+ "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
112
+ "model.layers.16.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
113
+ "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
114
+ "model.layers.16.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
115
+ "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
116
+ "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors",
117
+ "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
118
+ "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
119
+ "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
120
+ "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
121
+ "model.layers.17.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
122
+ "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
123
+ "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
124
+ "model.layers.17.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
125
+ "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
126
+ "model.layers.17.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
127
+ "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
128
+ "model.layers.18.input_layernorm.weight": "model-00003-of-00004.safetensors",
129
+ "model.layers.18.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
130
+ "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
131
+ "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
132
+ "model.layers.18.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
133
+ "model.layers.18.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
134
+ "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
135
+ "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
136
+ "model.layers.18.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
137
+ "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
138
+ "model.layers.18.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
139
+ "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
140
+ "model.layers.19.input_layernorm.weight": "model-00003-of-00004.safetensors",
141
+ "model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
142
+ "model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
143
+ "model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
144
+ "model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
145
+ "model.layers.19.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
146
+ "model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
147
+ "model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
148
+ "model.layers.19.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
149
+ "model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
150
+ "model.layers.19.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
151
+ "model.layers.19.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
152
+ "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
153
+ "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
154
+ "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
155
+ "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
156
+ "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
157
+ "model.layers.2.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
158
+ "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
159
+ "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
160
+ "model.layers.2.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
161
+ "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
162
+ "model.layers.2.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
163
+ "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
164
+ "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
165
+ "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
166
+ "model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
167
+ "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
168
+ "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
169
+ "model.layers.20.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
170
+ "model.layers.20.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
171
+ "model.layers.20.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
172
+ "model.layers.20.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
173
+ "model.layers.20.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
174
+ "model.layers.20.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
175
+ "model.layers.20.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
176
+ "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
177
+ "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
178
+ "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
179
+ "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
180
+ "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
181
+ "model.layers.21.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
182
+ "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
183
+ "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
184
+ "model.layers.21.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
185
+ "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
186
+ "model.layers.21.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
187
+ "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
188
+ "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
189
+ "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
190
+ "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
191
+ "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
192
+ "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
193
+ "model.layers.22.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
194
+ "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
195
+ "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
196
+ "model.layers.22.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
197
+ "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
198
+ "model.layers.22.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
199
+ "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
200
+ "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
201
+ "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
202
+ "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
203
+ "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
204
+ "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
205
+ "model.layers.23.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
206
+ "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
207
+ "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
208
+ "model.layers.23.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
209
+ "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
210
+ "model.layers.23.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
211
+ "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
212
+ "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
213
+ "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
214
+ "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
215
+ "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
216
+ "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
217
+ "model.layers.24.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
218
+ "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
219
+ "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
220
+ "model.layers.24.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
221
+ "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
222
+ "model.layers.24.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
223
+ "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
224
+ "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
225
+ "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
226
+ "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
227
+ "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
228
+ "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
229
+ "model.layers.25.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
230
+ "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
231
+ "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
232
+ "model.layers.25.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
233
+ "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
234
+ "model.layers.25.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
235
+ "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
236
+ "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors",
237
+ "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
238
+ "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
239
+ "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
240
+ "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
241
+ "model.layers.26.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
242
+ "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
243
+ "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
244
+ "model.layers.26.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
245
+ "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
246
+ "model.layers.26.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
247
+ "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
248
+ "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors",
249
+ "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
250
+ "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
251
+ "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
252
+ "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
253
+ "model.layers.27.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
254
+ "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
255
+ "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
256
+ "model.layers.27.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
257
+ "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
258
+ "model.layers.27.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
259
+ "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
260
+ "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors",
261
+ "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
262
+ "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
263
+ "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
264
+ "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
265
+ "model.layers.3.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
266
+ "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
267
+ "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
268
+ "model.layers.3.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
269
+ "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
270
+ "model.layers.3.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
271
+ "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
272
+ "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors",
273
+ "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
274
+ "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
275
+ "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
276
+ "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
277
+ "model.layers.4.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
278
+ "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
279
+ "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
280
+ "model.layers.4.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
281
+ "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
282
+ "model.layers.4.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
283
+ "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
284
+ "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors",
285
+ "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
286
+ "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
287
+ "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
288
+ "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
289
+ "model.layers.5.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
290
+ "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
291
+ "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
292
+ "model.layers.5.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
293
+ "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
294
+ "model.layers.5.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
295
+ "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
296
+ "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors",
297
+ "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
298
+ "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
299
+ "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
300
+ "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
301
+ "model.layers.6.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
302
+ "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
303
+ "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
304
+ "model.layers.6.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
305
+ "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
306
+ "model.layers.6.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
307
+ "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
308
+ "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors",
309
+ "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
310
+ "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
311
+ "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
312
+ "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
313
+ "model.layers.7.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
314
+ "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
315
+ "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
316
+ "model.layers.7.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
317
+ "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
318
+ "model.layers.7.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
319
+ "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
320
+ "model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
321
+ "model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
322
+ "model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
323
+ "model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
324
+ "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
325
+ "model.layers.8.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
326
+ "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
327
+ "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
328
+ "model.layers.8.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
329
+ "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
330
+ "model.layers.8.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
331
+ "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
332
+ "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
333
+ "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
334
+ "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
335
+ "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
336
+ "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
337
+ "model.layers.9.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
338
+ "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
339
+ "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
340
+ "model.layers.9.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
341
+ "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
342
+ "model.layers.9.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
343
+ "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
344
+ "model.norm.weight": "model-00003-of-00004.safetensors"
345
+ }
346
+ }
rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99597ef7442635c9fc7e33f58ab65a1f097883076cda723846953702c5b1bb41
3
+ size 15920
rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e9986a5b4d0c0854c31b0a552b36c8d61c4166799896370911745287d59eb2f1
3
+ size 15920
rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62d0a03a6899b1d9c6471730a5f27a61c6d5decea4fa364e33de7339d74afa6a
3
+ size 15920
rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5d2deda76dfce536554e9a353390020e1434a02190925740d534a1f14df2db5
3
+ size 15920
rng_state_4.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ea03f5389460add8ac4bb962ff352abda89dc8283120a1b75a0b08a8701c3d0
3
+ size 15920
rng_state_5.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d0efd6bdda2dc5f3a2dc45fe493b81426aefc76e034baaf43d92cd400d55abbf
3
+ size 15920
rng_state_6.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:15abd0cbcd121c12ef3480b67be50888f282d5deb1f190c5739144faa876b79e
3
+ size 15920
rng_state_7.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dfbfa4971e4f22128a2a3938f9e09ee46a74b13ba2c990c12f6badfb97ceb345
3
+ size 15920
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8fb3eaf688658712768ae645e4a2b5f778a59201f17641a26e210365d8c2ef6f
3
+ size 1064
special_tokens_map.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<｜begin▁of▁sentence｜>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<｜end▁of▁sentence｜>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<｜end▁of▁sentence｜>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ }
23
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e20ddafc659ba90242154b55275402edeca0715e5dbb30f56815a4ce081f4893
3
+ size 11422778
tokenizer_config.json ADDED
@@ -0,0 +1,195 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": null,
5
+ "added_tokens_decoder": {
6
+ "151643": {
7
+ "content": "<｜end▁of▁sentence｜>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "151644": {
15
+ "content": "<｜User｜>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": false
21
+ },
22
+ "151645": {
23
+ "content": "<｜Assistant｜>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": false
29
+ },
30
+ "151646": {
31
+ "content": "<｜begin▁of▁sentence｜>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false,
36
+ "special": true
37
+ },
38
+ "151647": {
39
+ "content": "<|EOT|>",
40
+ "lstrip": false,
41
+ "normalized": false,
42
+ "rstrip": false,
43
+ "single_word": false,
44
+ "special": false
45
+ },
46
+ "151648": {
47
+ "content": "<think>",
48
+ "lstrip": false,
49
+ "normalized": false,
50
+ "rstrip": false,
51
+ "single_word": false,
52
+ "special": false
53
+ },
54
+ "151649": {
55
+ "content": "</think>",
56
+ "lstrip": false,
57
+ "normalized": false,
58
+ "rstrip": false,
59
+ "single_word": false,
60
+ "special": false
61
+ },
62
+ "151650": {
63
+ "content": "<|quad_start|>",
64
+ "lstrip": false,
65
+ "normalized": false,
66
+ "rstrip": false,
67
+ "single_word": false,
68
+ "special": true
69
+ },
70
+ "151651": {
71
+ "content": "<|quad_end|>",
72
+ "lstrip": false,
73
+ "normalized": false,
74
+ "rstrip": false,
75
+ "single_word": false,
76
+ "special": true
77
+ },
78
+ "151652": {
79
+ "content": "<|vision_start|>",
80
+ "lstrip": false,
81
+ "normalized": false,
82
+ "rstrip": false,
83
+ "single_word": false,
84
+ "special": true
85
+ },
86
+ "151653": {
87
+ "content": "<|vision_end|>",
88
+ "lstrip": false,
89
+ "normalized": false,
90
+ "rstrip": false,
91
+ "single_word": false,
92
+ "special": true
93
+ },
94
+ "151654": {
95
+ "content": "<|vision_pad|>",
96
+ "lstrip": false,
97
+ "normalized": false,
98
+ "rstrip": false,
99
+ "single_word": false,
100
+ "special": true
101
+ },
102
+ "151655": {
103
+ "content": "<|image_pad|>",
104
+ "lstrip": false,
105
+ "normalized": false,
106
+ "rstrip": false,
107
+ "single_word": false,
108
+ "special": true
109
+ },
110
+ "151656": {
111
+ "content": "<|video_pad|>",
112
+ "lstrip": false,
113
+ "normalized": false,
114
+ "rstrip": false,
115
+ "single_word": false,
116
+ "special": true
117
+ },
118
+ "151657": {
119
+ "content": "<tool_call>",
120
+ "lstrip": false,
121
+ "normalized": false,
122
+ "rstrip": false,
123
+ "single_word": false,
124
+ "special": false
125
+ },
126
+ "151658": {
127
+ "content": "</tool_call>",
128
+ "lstrip": false,
129
+ "normalized": false,
130
+ "rstrip": false,
131
+ "single_word": false,
132
+ "special": false
133
+ },
134
+ "151659": {
135
+ "content": "<|fim_prefix|>",
136
+ "lstrip": false,
137
+ "normalized": false,
138
+ "rstrip": false,
139
+ "single_word": false,
140
+ "special": false
141
+ },
142
+ "151660": {
143
+ "content": "<|fim_middle|>",
144
+ "lstrip": false,
145
+ "normalized": false,
146
+ "rstrip": false,
147
+ "single_word": false,
148
+ "special": false
149
+ },
150
+ "151661": {
151
+ "content": "<|fim_suffix|>",
152
+ "lstrip": false,
153
+ "normalized": false,
154
+ "rstrip": false,
155
+ "single_word": false,
156
+ "special": false
157
+ },
158
+ "151662": {
159
+ "content": "<|fim_pad|>",
160
+ "lstrip": false,
161
+ "normalized": false,
162
+ "rstrip": false,
163
+ "single_word": false,
164
+ "special": false
165
+ },
166
+ "151663": {
167
+ "content": "<|repo_name|>",
168
+ "lstrip": false,
169
+ "normalized": false,
170
+ "rstrip": false,
171
+ "single_word": false,
172
+ "special": false
173
+ },
174
+ "151664": {
175
+ "content": "<|file_sep|>",
176
+ "lstrip": false,
177
+ "normalized": false,
178
+ "rstrip": false,
179
+ "single_word": false,
180
+ "special": false
181
+ }
182
+ },
183
+ "bos_token": "<｜begin▁of▁sentence｜>",
184
+ "chat_template": "{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% set ns = namespace(is_first=false, is_tool=false, is_output_first=true, system_prompt='') %}{%- for message in messages %}{%- if message['role'] == 'system' %}{% set ns.system_prompt = message['content'] %}{%- endif %}{%- endfor %}{{bos_token}}{{ns.system_prompt}}{%- for message in messages %}{%- if message['role'] == 'user' %}{%- set ns.is_tool = false -%}{{'<｜User｜>' + message['content']}}{%- endif %}{%- if message['role'] == 'assistant' and message['content'] is none %}{%- set ns.is_tool = false -%}{%- for tool in message['tool_calls']%}{%- if not ns.is_first %}{{'<｜Assistant｜><｜tool▁calls▁begin｜><｜tool▁call▁begin��>' + tool['type'] + '<｜tool▁sep｜>' + tool['function']['name'] + '\\n' + '```json' + '\\n' + tool['function']['arguments'] + '\\n' + '```' + '<｜tool▁call▁end｜>'}}{%- set ns.is_first = true -%}{%- else %}{{'\\n' + '<｜tool▁call▁begin｜>' + tool['type'] + '<｜tool▁sep｜>' + tool['function']['name'] + '\\n' + '```json' + '\\n' + tool['function']['arguments'] + '\\n' + '```' + '<｜tool▁call▁end｜>'}}{{'<｜tool▁calls▁end｜><｜end▁of▁sentence｜>'}}{%- endif %}{%- endfor %}{%- endif %}{%- if message['role'] == 'assistant' and message['content'] is not none %}{%- if ns.is_tool %}{{'<｜tool▁outputs▁end｜>' + message['content'] + '<｜end▁of▁sentence｜>'}}{%- set ns.is_tool = false -%}{%- else %}{% set content = message['content'] %}{% if '</think>' in content %}{% set content = content.split('</think>')[-1] %}{% endif %}{{'<｜Assistant｜>' + content + '<｜end▁of▁sentence｜>'}}{%- endif %}{%- endif %}{%- if message['role'] == 'tool' %}{%- set ns.is_tool = true -%}{%- if ns.is_output_first %}{{'<｜tool▁outputs▁begin｜><｜tool▁output▁begin｜>' + message['content'] + '<｜tool▁output▁end｜>'}}{%- set ns.is_output_first = false %}{%- else %}{{'\\n<｜tool▁output▁begin｜>' + message['content'] + '<｜tool▁output▁end｜>'}}{%- endif %}{%- endif %}{%- endfor -%}{% if ns.is_tool %}{{'<｜tool▁outputs▁end｜>'}}{% endif %}{% if add_generation_prompt and not ns.is_tool %}{{'<｜Assistant｜><think>\\n'}}{% endif %}",
185
+ "clean_up_tokenization_spaces": false,
186
+ "eos_token": "<｜end▁of▁sentence｜>",
187
+ "legacy": true,
188
+ "model_max_length": 8192,
189
+ "pad_token": "<｜end▁of▁sentence｜>",
190
+ "padding_side": "left",
191
+ "sp_model_kwargs": {},
192
+ "tokenizer_class": "LlamaTokenizer",
193
+ "unk_token": null,
194
+ "use_default_system_prompt": false
195
+ }
trainer_state.json ADDED
@@ -0,0 +1,2113 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 8,
6
+ "global_step": 256,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.00390625,
13
+ "grad_norm": 3.380525042530954,
14
+ "learning_rate": 1e-05,
15
+ "loss": 0.2859,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.0078125,
20
+ "grad_norm": 2.6901012326349156,
21
+ "learning_rate": 1e-05,
22
+ "loss": 0.2117,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.01171875,
27
+ "grad_norm": 3.191447237922227,
28
+ "learning_rate": 1e-05,
29
+ "loss": 0.2602,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.015625,
34
+ "grad_norm": 2.204083519446381,
35
+ "learning_rate": 1e-05,
36
+ "loss": 0.1972,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.01953125,
41
+ "grad_norm": 2.0481149317155687,
42
+ "learning_rate": 1e-05,
43
+ "loss": 0.2338,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.0234375,
48
+ "grad_norm": 1.6269814174466988,
49
+ "learning_rate": 1e-05,
50
+ "loss": 0.214,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.02734375,
55
+ "grad_norm": 1.6908703624878527,
56
+ "learning_rate": 1e-05,
57
+ "loss": 0.2088,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.03125,
62
+ "grad_norm": 1.2059719622160197,
63
+ "learning_rate": 1e-05,
64
+ "loss": 0.1975,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.03125,
69
+ "eval_dev_acc": 0.515625,
70
+ "eval_dev_token": 4849.7578125,
71
+ "eval_runtime": 168.4394,
72
+ "eval_samples_per_second": 0.095,
73
+ "eval_steps_per_second": 0.006,
74
+ "step": 8
75
+ },
76
+ {
77
+ "epoch": 0.03515625,
78
+ "grad_norm": 1.6837720712641369,
79
+ "learning_rate": 1e-05,
80
+ "loss": 0.1873,
81
+ "step": 9
82
+ },
83
+ {
84
+ "epoch": 0.0390625,
85
+ "grad_norm": 1.2504651087103098,
86
+ "learning_rate": 1e-05,
87
+ "loss": 0.1959,
88
+ "step": 10
89
+ },
90
+ {
91
+ "epoch": 0.04296875,
92
+ "grad_norm": 1.3187603751382884,
93
+ "learning_rate": 1e-05,
94
+ "loss": 0.2135,
95
+ "step": 11
96
+ },
97
+ {
98
+ "epoch": 0.046875,
99
+ "grad_norm": 1.3545446581007174,
100
+ "learning_rate": 1e-05,
101
+ "loss": 0.2428,
102
+ "step": 12
103
+ },
104
+ {
105
+ "epoch": 0.05078125,
106
+ "grad_norm": 1.6286051945906104,
107
+ "learning_rate": 1e-05,
108
+ "loss": 0.1708,
109
+ "step": 13
110
+ },
111
+ {
112
+ "epoch": 0.0546875,
113
+ "grad_norm": 1.6081830921647842,
114
+ "learning_rate": 1e-05,
115
+ "loss": 0.1471,
116
+ "step": 14
117
+ },
118
+ {
119
+ "epoch": 0.05859375,
120
+ "grad_norm": 1.4305460955933824,
121
+ "learning_rate": 1e-05,
122
+ "loss": 0.1837,
123
+ "step": 15
124
+ },
125
+ {
126
+ "epoch": 0.0625,
127
+ "grad_norm": 1.3961670104174644,
128
+ "learning_rate": 1e-05,
129
+ "loss": 0.1352,
130
+ "step": 16
131
+ },
132
+ {
133
+ "epoch": 0.0625,
134
+ "eval_dev_acc": 0.4296875,
135
+ "eval_dev_token": 5067.265625,
136
+ "eval_runtime": 167.2848,
137
+ "eval_samples_per_second": 0.096,
138
+ "eval_steps_per_second": 0.006,
139
+ "step": 16
140
+ },
141
+ {
142
+ "epoch": 0.06640625,
143
+ "grad_norm": 1.5507019702345457,
144
+ "learning_rate": 1e-05,
145
+ "loss": 0.1657,
146
+ "step": 17
147
+ },
148
+ {
149
+ "epoch": 0.0703125,
150
+ "grad_norm": 1.3395286968352729,
151
+ "learning_rate": 1e-05,
152
+ "loss": 0.1824,
153
+ "step": 18
154
+ },
155
+ {
156
+ "epoch": 0.07421875,
157
+ "grad_norm": 2.201219146342779,
158
+ "learning_rate": 1e-05,
159
+ "loss": 0.1391,
160
+ "step": 19
161
+ },
162
+ {
163
+ "epoch": 0.078125,
164
+ "grad_norm": 1.75559779570709,
165
+ "learning_rate": 1e-05,
166
+ "loss": 0.1351,
167
+ "step": 20
168
+ },
169
+ {
170
+ "epoch": 0.08203125,
171
+ "grad_norm": 2.0359121335172428,
172
+ "learning_rate": 1e-05,
173
+ "loss": 0.1748,
174
+ "step": 21
175
+ },
176
+ {
177
+ "epoch": 0.0859375,
178
+ "grad_norm": 1.6822343317370052,
179
+ "learning_rate": 1e-05,
180
+ "loss": 0.1582,
181
+ "step": 22
182
+ },
183
+ {
184
+ "epoch": 0.08984375,
185
+ "grad_norm": 1.9664935447837442,
186
+ "learning_rate": 1e-05,
187
+ "loss": 0.1338,
188
+ "step": 23
189
+ },
190
+ {
191
+ "epoch": 0.09375,
192
+ "grad_norm": 1.1463903797363937,
193
+ "learning_rate": 1e-05,
194
+ "loss": 0.1139,
195
+ "step": 24
196
+ },
197
+ {
198
+ "epoch": 0.09375,
199
+ "eval_dev_acc": 0.4296875,
200
+ "eval_dev_token": 4994.296875,
201
+ "eval_runtime": 168.4043,
202
+ "eval_samples_per_second": 0.095,
203
+ "eval_steps_per_second": 0.006,
204
+ "step": 24
205
+ },
206
+ {
207
+ "epoch": 0.09765625,
208
+ "grad_norm": 2.1728621095149627,
209
+ "learning_rate": 1e-05,
210
+ "loss": 0.1471,
211
+ "step": 25
212
+ },
213
+ {
214
+ "epoch": 0.1015625,
215
+ "grad_norm": 1.6714738223766954,
216
+ "learning_rate": 1e-05,
217
+ "loss": 0.1349,
218
+ "step": 26
219
+ },
220
+ {
221
+ "epoch": 0.10546875,
222
+ "grad_norm": 1.5574316583381629,
223
+ "learning_rate": 1e-05,
224
+ "loss": 0.1356,
225
+ "step": 27
226
+ },
227
+ {
228
+ "epoch": 0.109375,
229
+ "grad_norm": 1.4728847084572547,
230
+ "learning_rate": 1e-05,
231
+ "loss": 0.1509,
232
+ "step": 28
233
+ },
234
+ {
235
+ "epoch": 0.11328125,
236
+ "grad_norm": 1.4769394661942852,
237
+ "learning_rate": 1e-05,
238
+ "loss": 0.1294,
239
+ "step": 29
240
+ },
241
+ {
242
+ "epoch": 0.1171875,
243
+ "grad_norm": 1.8550097520759188,
244
+ "learning_rate": 1e-05,
245
+ "loss": 0.1208,
246
+ "step": 30
247
+ },
248
+ {
249
+ "epoch": 0.12109375,
250
+ "grad_norm": 1.75157088447911,
251
+ "learning_rate": 1e-05,
252
+ "loss": 0.0993,
253
+ "step": 31
254
+ },
255
+ {
256
+ "epoch": 0.125,
257
+ "grad_norm": 1.6233472727407252,
258
+ "learning_rate": 1e-05,
259
+ "loss": 0.1412,
260
+ "step": 32
261
+ },
262
+ {
263
+ "epoch": 0.125,
264
+ "eval_dev_acc": 0.4609375,
265
+ "eval_dev_token": 4228.15625,
266
+ "eval_runtime": 159.0398,
267
+ "eval_samples_per_second": 0.101,
268
+ "eval_steps_per_second": 0.006,
269
+ "step": 32
270
+ },
271
+ {
272
+ "epoch": 0.12890625,
273
+ "grad_norm": 1.5246001678514782,
274
+ "learning_rate": 1e-05,
275
+ "loss": 0.1268,
276
+ "step": 33
277
+ },
278
+ {
279
+ "epoch": 0.1328125,
280
+ "grad_norm": 1.020147996755851,
281
+ "learning_rate": 1e-05,
282
+ "loss": 0.166,
283
+ "step": 34
284
+ },
285
+ {
286
+ "epoch": 0.13671875,
287
+ "grad_norm": 0.9795032964583498,
288
+ "learning_rate": 1e-05,
289
+ "loss": 0.1223,
290
+ "step": 35
291
+ },
292
+ {
293
+ "epoch": 0.140625,
294
+ "grad_norm": 1.0328587053324862,
295
+ "learning_rate": 1e-05,
296
+ "loss": 0.0889,
297
+ "step": 36
298
+ },
299
+ {
300
+ "epoch": 0.14453125,
301
+ "grad_norm": 0.8587530858129762,
302
+ "learning_rate": 1e-05,
303
+ "loss": 0.1618,
304
+ "step": 37
305
+ },
306
+ {
307
+ "epoch": 0.1484375,
308
+ "grad_norm": 1.0451234874371433,
309
+ "learning_rate": 1e-05,
310
+ "loss": 0.1973,
311
+ "step": 38
312
+ },
313
+ {
314
+ "epoch": 0.15234375,
315
+ "grad_norm": 1.032741287831154,
316
+ "learning_rate": 1e-05,
317
+ "loss": 0.1999,
318
+ "step": 39
319
+ },
320
+ {
321
+ "epoch": 0.15625,
322
+ "grad_norm": 1.0128010813738295,
323
+ "learning_rate": 1e-05,
324
+ "loss": 0.1314,
325
+ "step": 40
326
+ },
327
+ {
328
+ "epoch": 0.15625,
329
+ "eval_dev_acc": 0.40625,
330
+ "eval_dev_token": 5015.7421875,
331
+ "eval_runtime": 167.9354,
332
+ "eval_samples_per_second": 0.095,
333
+ "eval_steps_per_second": 0.006,
334
+ "step": 40
335
+ },
336
+ {
337
+ "epoch": 0.16015625,
338
+ "grad_norm": 0.7085331860395175,
339
+ "learning_rate": 1e-05,
340
+ "loss": 0.1424,
341
+ "step": 41
342
+ },
343
+ {
344
+ "epoch": 0.1640625,
345
+ "grad_norm": 0.8522197113830303,
346
+ "learning_rate": 1e-05,
347
+ "loss": 0.1523,
348
+ "step": 42
349
+ },
350
+ {
351
+ "epoch": 0.16796875,
352
+ "grad_norm": 0.9700458234990689,
353
+ "learning_rate": 1e-05,
354
+ "loss": 0.1655,
355
+ "step": 43
356
+ },
357
+ {
358
+ "epoch": 0.171875,
359
+ "grad_norm": 2.0713947251278855,
360
+ "learning_rate": 1e-05,
361
+ "loss": 0.2946,
362
+ "step": 44
363
+ },
364
+ {
365
+ "epoch": 0.17578125,
366
+ "grad_norm": 1.6441862242379885,
367
+ "learning_rate": 1e-05,
368
+ "loss": 0.2547,
369
+ "step": 45
370
+ },
371
+ {
372
+ "epoch": 0.1796875,
373
+ "grad_norm": 1.7959964112861366,
374
+ "learning_rate": 1e-05,
375
+ "loss": 0.3009,
376
+ "step": 46
377
+ },
378
+ {
379
+ "epoch": 0.18359375,
380
+ "grad_norm": 1.3449858551505456,
381
+ "learning_rate": 1e-05,
382
+ "loss": 0.2094,
383
+ "step": 47
384
+ },
385
+ {
386
+ "epoch": 0.1875,
387
+ "grad_norm": 1.2087309569022056,
388
+ "learning_rate": 1e-05,
389
+ "loss": 0.1908,
390
+ "step": 48
391
+ },
392
+ {
393
+ "epoch": 0.1875,
394
+ "eval_dev_acc": 0.34375,
395
+ "eval_dev_token": 4538.84375,
396
+ "eval_runtime": 161.6976,
397
+ "eval_samples_per_second": 0.099,
398
+ "eval_steps_per_second": 0.006,
399
+ "step": 48
400
+ },
401
+ {
402
+ "epoch": 0.19140625,
403
+ "grad_norm": 1.1559146316352948,
404
+ "learning_rate": 1e-05,
405
+ "loss": 0.3036,
406
+ "step": 49
407
+ },
408
+ {
409
+ "epoch": 0.1953125,
410
+ "grad_norm": 1.131769529502962,
411
+ "learning_rate": 1e-05,
412
+ "loss": 0.2441,
413
+ "step": 50
414
+ },
415
+ {
416
+ "epoch": 0.19921875,
417
+ "grad_norm": 1.4116452844735226,
418
+ "learning_rate": 1e-05,
419
+ "loss": 0.2028,
420
+ "step": 51
421
+ },
422
+ {
423
+ "epoch": 0.203125,
424
+ "grad_norm": 0.7550364491986332,
425
+ "learning_rate": 1e-05,
426
+ "loss": 0.215,
427
+ "step": 52
428
+ },
429
+ {
430
+ "epoch": 0.20703125,
431
+ "grad_norm": 1.3915284765850489,
432
+ "learning_rate": 1e-05,
433
+ "loss": 0.2878,
434
+ "step": 53
435
+ },
436
+ {
437
+ "epoch": 0.2109375,
438
+ "grad_norm": 1.6351241901381652,
439
+ "learning_rate": 1e-05,
440
+ "loss": 0.2446,
441
+ "step": 54
442
+ },
443
+ {
444
+ "epoch": 0.21484375,
445
+ "grad_norm": 1.6083218458029132,
446
+ "learning_rate": 1e-05,
447
+ "loss": 0.2088,
448
+ "step": 55
449
+ },
450
+ {
451
+ "epoch": 0.21875,
452
+ "grad_norm": 0.7434150303822764,
453
+ "learning_rate": 1e-05,
454
+ "loss": 0.2262,
455
+ "step": 56
456
+ },
457
+ {
458
+ "epoch": 0.21875,
459
+ "eval_dev_acc": 0.30708661675453186,
460
+ "eval_dev_token": 5670.251953125,
461
+ "eval_runtime": 174.7692,
462
+ "eval_samples_per_second": 0.092,
463
+ "eval_steps_per_second": 0.006,
464
+ "step": 56
465
+ },
466
+ {
467
+ "epoch": 0.22265625,
468
+ "grad_norm": 1.0769799759099778,
469
+ "learning_rate": 1e-05,
470
+ "loss": 0.208,
471
+ "step": 57
472
+ },
473
+ {
474
+ "epoch": 0.2265625,
475
+ "grad_norm": 0.9298141621627772,
476
+ "learning_rate": 1e-05,
477
+ "loss": 0.1687,
478
+ "step": 58
479
+ },
480
+ {
481
+ "epoch": 0.23046875,
482
+ "grad_norm": 1.285492123129724,
483
+ "learning_rate": 1e-05,
484
+ "loss": 0.2427,
485
+ "step": 59
486
+ },
487
+ {
488
+ "epoch": 0.234375,
489
+ "grad_norm": 0.8346778861730894,
490
+ "learning_rate": 1e-05,
491
+ "loss": 0.219,
492
+ "step": 60
493
+ },
494
+ {
495
+ "epoch": 0.23828125,
496
+ "grad_norm": 0.9873196942775492,
497
+ "learning_rate": 1e-05,
498
+ "loss": 0.242,
499
+ "step": 61
500
+ },
501
+ {
502
+ "epoch": 0.2421875,
503
+ "grad_norm": 0.9596507860915271,
504
+ "learning_rate": 1e-05,
505
+ "loss": 0.2148,
506
+ "step": 62
507
+ },
508
+ {
509
+ "epoch": 0.24609375,
510
+ "grad_norm": 1.0988562593647762,
511
+ "learning_rate": 1e-05,
512
+ "loss": 0.2396,
513
+ "step": 63
514
+ },
515
+ {
516
+ "epoch": 0.25,
517
+ "grad_norm": 0.9707635131928222,
518
+ "learning_rate": 1e-05,
519
+ "loss": 0.238,
520
+ "step": 64
521
+ },
522
+ {
523
+ "epoch": 0.25,
524
+ "eval_dev_acc": 0.5390625,
525
+ "eval_dev_token": 4394.921875,
526
+ "eval_runtime": 161.3481,
527
+ "eval_samples_per_second": 0.099,
528
+ "eval_steps_per_second": 0.006,
529
+ "step": 64
530
+ },
531
+ {
532
+ "epoch": 0.25390625,
533
+ "grad_norm": 0.8083595053544823,
534
+ "learning_rate": 1e-05,
535
+ "loss": 0.293,
536
+ "step": 65
537
+ },
538
+ {
539
+ "epoch": 0.2578125,
540
+ "grad_norm": 0.6893947679382126,
541
+ "learning_rate": 1e-05,
542
+ "loss": 0.2866,
543
+ "step": 66
544
+ },
545
+ {
546
+ "epoch": 0.26171875,
547
+ "grad_norm": 1.0271679359276198,
548
+ "learning_rate": 1e-05,
549
+ "loss": 0.2276,
550
+ "step": 67
551
+ },
552
+ {
553
+ "epoch": 0.265625,
554
+ "grad_norm": 1.1776528602190077,
555
+ "learning_rate": 1e-05,
556
+ "loss": 0.1887,
557
+ "step": 68
558
+ },
559
+ {
560
+ "epoch": 0.26953125,
561
+ "grad_norm": 1.163717423684938,
562
+ "learning_rate": 1e-05,
563
+ "loss": 0.2147,
564
+ "step": 69
565
+ },
566
+ {
567
+ "epoch": 0.2734375,
568
+ "grad_norm": 0.8134427746893115,
569
+ "learning_rate": 1e-05,
570
+ "loss": 0.2342,
571
+ "step": 70
572
+ },
573
+ {
574
+ "epoch": 0.27734375,
575
+ "grad_norm": 1.4269332848478926,
576
+ "learning_rate": 1e-05,
577
+ "loss": 0.1919,
578
+ "step": 71
579
+ },
580
+ {
581
+ "epoch": 0.28125,
582
+ "grad_norm": 0.8200789264174901,
583
+ "learning_rate": 1e-05,
584
+ "loss": 0.2175,
585
+ "step": 72
586
+ },
587
+ {
588
+ "epoch": 0.28125,
589
+ "eval_dev_acc": 0.53125,
590
+ "eval_dev_token": 4859.7421875,
591
+ "eval_runtime": 166.6197,
592
+ "eval_samples_per_second": 0.096,
593
+ "eval_steps_per_second": 0.006,
594
+ "step": 72
595
+ },
596
+ {
597
+ "epoch": 0.28515625,
598
+ "grad_norm": 1.007316679088458,
599
+ "learning_rate": 1e-05,
600
+ "loss": 0.3108,
601
+ "step": 73
602
+ },
603
+ {
604
+ "epoch": 0.2890625,
605
+ "grad_norm": 0.6637709768510952,
606
+ "learning_rate": 1e-05,
607
+ "loss": 0.1794,
608
+ "step": 74
609
+ },
610
+ {
611
+ "epoch": 0.29296875,
612
+ "grad_norm": 1.0144512803754202,
613
+ "learning_rate": 1e-05,
614
+ "loss": 0.1905,
615
+ "step": 75
616
+ },
617
+ {
618
+ "epoch": 0.296875,
619
+ "grad_norm": 1.2499777112248354,
620
+ "learning_rate": 1e-05,
621
+ "loss": 0.2014,
622
+ "step": 76
623
+ },
624
+ {
625
+ "epoch": 0.30078125,
626
+ "grad_norm": 1.0642239482819718,
627
+ "learning_rate": 1e-05,
628
+ "loss": 0.1648,
629
+ "step": 77
630
+ },
631
+ {
632
+ "epoch": 0.3046875,
633
+ "grad_norm": 0.8739614674360524,
634
+ "learning_rate": 1e-05,
635
+ "loss": 0.1537,
636
+ "step": 78
637
+ },
638
+ {
639
+ "epoch": 0.30859375,
640
+ "grad_norm": 0.5320613340314281,
641
+ "learning_rate": 1e-05,
642
+ "loss": 0.2128,
643
+ "step": 79
644
+ },
645
+ {
646
+ "epoch": 0.3125,
647
+ "grad_norm": 1.2802208673828028,
648
+ "learning_rate": 1e-05,
649
+ "loss": 0.1939,
650
+ "step": 80
651
+ },
652
+ {
653
+ "epoch": 0.3125,
654
+ "eval_dev_acc": 0.4609375,
655
+ "eval_dev_token": 5065.421875,
656
+ "eval_runtime": 168.4523,
657
+ "eval_samples_per_second": 0.095,
658
+ "eval_steps_per_second": 0.006,
659
+ "step": 80
660
+ },
661
+ {
662
+ "epoch": 0.31640625,
663
+ "grad_norm": 1.1564057868614226,
664
+ "learning_rate": 1e-05,
665
+ "loss": 0.2215,
666
+ "step": 81
667
+ },
668
+ {
669
+ "epoch": 0.3203125,
670
+ "grad_norm": 0.7104999594850884,
671
+ "learning_rate": 1e-05,
672
+ "loss": 0.1224,
673
+ "step": 82
674
+ },
675
+ {
676
+ "epoch": 0.32421875,
677
+ "grad_norm": 0.6466657594813067,
678
+ "learning_rate": 1e-05,
679
+ "loss": 0.145,
680
+ "step": 83
681
+ },
682
+ {
683
+ "epoch": 0.328125,
684
+ "grad_norm": 1.3499118701284736,
685
+ "learning_rate": 1e-05,
686
+ "loss": 0.1963,
687
+ "step": 84
688
+ },
689
+ {
690
+ "epoch": 0.33203125,
691
+ "grad_norm": 0.6363338361760021,
692
+ "learning_rate": 1e-05,
693
+ "loss": 0.1781,
694
+ "step": 85
695
+ },
696
+ {
697
+ "epoch": 0.3359375,
698
+ "grad_norm": 0.8807906150832371,
699
+ "learning_rate": 1e-05,
700
+ "loss": 0.1426,
701
+ "step": 86
702
+ },
703
+ {
704
+ "epoch": 0.33984375,
705
+ "grad_norm": 0.7466707582875238,
706
+ "learning_rate": 1e-05,
707
+ "loss": 0.1629,
708
+ "step": 87
709
+ },
710
+ {
711
+ "epoch": 0.34375,
712
+ "grad_norm": 0.7773292125565866,
713
+ "learning_rate": 1e-05,
714
+ "loss": 0.181,
715
+ "step": 88
716
+ },
717
+ {
718
+ "epoch": 0.34375,
719
+ "eval_dev_acc": 0.4609375,
720
+ "eval_dev_token": 5092.8984375,
721
+ "eval_runtime": 168.9275,
722
+ "eval_samples_per_second": 0.095,
723
+ "eval_steps_per_second": 0.006,
724
+ "step": 88
725
+ },
726
+ {
727
+ "epoch": 0.34765625,
728
+ "grad_norm": 0.9798290139606278,
729
+ "learning_rate": 1e-05,
730
+ "loss": 0.1725,
731
+ "step": 89
732
+ },
733
+ {
734
+ "epoch": 0.3515625,
735
+ "grad_norm": 1.2761428002675261,
736
+ "learning_rate": 1e-05,
737
+ "loss": 0.175,
738
+ "step": 90
739
+ },
740
+ {
741
+ "epoch": 0.35546875,
742
+ "grad_norm": 0.5042091805859357,
743
+ "learning_rate": 1e-05,
744
+ "loss": 0.218,
745
+ "step": 91
746
+ },
747
+ {
748
+ "epoch": 0.359375,
749
+ "grad_norm": 1.017358230975041,
750
+ "learning_rate": 1e-05,
751
+ "loss": 0.2502,
752
+ "step": 92
753
+ },
754
+ {
755
+ "epoch": 0.36328125,
756
+ "grad_norm": 0.7366049175316091,
757
+ "learning_rate": 1e-05,
758
+ "loss": 0.1656,
759
+ "step": 93
760
+ },
761
+ {
762
+ "epoch": 0.3671875,
763
+ "grad_norm": 0.9422427666318486,
764
+ "learning_rate": 1e-05,
765
+ "loss": 0.1455,
766
+ "step": 94
767
+ },
768
+ {
769
+ "epoch": 0.37109375,
770
+ "grad_norm": 0.7689775552730859,
771
+ "learning_rate": 1e-05,
772
+ "loss": 0.1485,
773
+ "step": 95
774
+ },
775
+ {
776
+ "epoch": 0.375,
777
+ "grad_norm": 0.9090457524355386,
778
+ "learning_rate": 1e-05,
779
+ "loss": 0.1411,
780
+ "step": 96
781
+ },
782
+ {
783
+ "epoch": 0.375,
784
+ "eval_dev_acc": 0.453125,
785
+ "eval_dev_token": 4948.8359375,
786
+ "eval_runtime": 165.5377,
787
+ "eval_samples_per_second": 0.097,
788
+ "eval_steps_per_second": 0.006,
789
+ "step": 96
790
+ },
791
+ {
792
+ "epoch": 0.37890625,
793
+ "grad_norm": 0.7235724828873173,
794
+ "learning_rate": 1e-05,
795
+ "loss": 0.2193,
796
+ "step": 97
797
+ },
798
+ {
799
+ "epoch": 0.3828125,
800
+ "grad_norm": 0.7200445685294068,
801
+ "learning_rate": 1e-05,
802
+ "loss": 0.1985,
803
+ "step": 98
804
+ },
805
+ {
806
+ "epoch": 0.38671875,
807
+ "grad_norm": 0.6060156821220763,
808
+ "learning_rate": 1e-05,
809
+ "loss": 0.2096,
810
+ "step": 99
811
+ },
812
+ {
813
+ "epoch": 0.390625,
814
+ "grad_norm": 0.7114968462244617,
815
+ "learning_rate": 1e-05,
816
+ "loss": 0.1928,
817
+ "step": 100
818
+ },
819
+ {
820
+ "epoch": 0.39453125,
821
+ "grad_norm": 0.6397518359548336,
822
+ "learning_rate": 1e-05,
823
+ "loss": 0.2165,
824
+ "step": 101
825
+ },
826
+ {
827
+ "epoch": 0.3984375,
828
+ "grad_norm": 0.7027126137819094,
829
+ "learning_rate": 1e-05,
830
+ "loss": 0.2263,
831
+ "step": 102
832
+ },
833
+ {
834
+ "epoch": 0.40234375,
835
+ "grad_norm": 0.8648981933002193,
836
+ "learning_rate": 1e-05,
837
+ "loss": 0.2874,
838
+ "step": 103
839
+ },
840
+ {
841
+ "epoch": 0.40625,
842
+ "grad_norm": 0.9742992968412495,
843
+ "learning_rate": 1e-05,
844
+ "loss": 0.1755,
845
+ "step": 104
846
+ },
847
+ {
848
+ "epoch": 0.40625,
849
+ "eval_dev_acc": 0.3515625,
850
+ "eval_dev_token": 5303.1796875,
851
+ "eval_runtime": 173.9477,
852
+ "eval_samples_per_second": 0.092,
853
+ "eval_steps_per_second": 0.006,
854
+ "step": 104
855
+ },
856
+ {
857
+ "epoch": 0.41015625,
858
+ "grad_norm": 0.6358933759276069,
859
+ "learning_rate": 1e-05,
860
+ "loss": 0.1907,
861
+ "step": 105
862
+ },
863
+ {
864
+ "epoch": 0.4140625,
865
+ "grad_norm": 0.7859972506268991,
866
+ "learning_rate": 1e-05,
867
+ "loss": 0.1731,
868
+ "step": 106
869
+ },
870
+ {
871
+ "epoch": 0.41796875,
872
+ "grad_norm": 0.6429885607052577,
873
+ "learning_rate": 1e-05,
874
+ "loss": 0.187,
875
+ "step": 107
876
+ },
877
+ {
878
+ "epoch": 0.421875,
879
+ "grad_norm": 0.6314004528855494,
880
+ "learning_rate": 1e-05,
881
+ "loss": 0.2185,
882
+ "step": 108
883
+ },
884
+ {
885
+ "epoch": 0.42578125,
886
+ "grad_norm": 0.8243656111706104,
887
+ "learning_rate": 1e-05,
888
+ "loss": 0.1384,
889
+ "step": 109
890
+ },
891
+ {
892
+ "epoch": 0.4296875,
893
+ "grad_norm": 0.7310074535827911,
894
+ "learning_rate": 1e-05,
895
+ "loss": 0.1724,
896
+ "step": 110
897
+ },
898
+ {
899
+ "epoch": 0.43359375,
900
+ "grad_norm": 1.8710293554497974,
901
+ "learning_rate": 1e-05,
902
+ "loss": 0.273,
903
+ "step": 111
904
+ },
905
+ {
906
+ "epoch": 0.4375,
907
+ "grad_norm": 1.3308164398688347,
908
+ "learning_rate": 1e-05,
909
+ "loss": 0.2852,
910
+ "step": 112
911
+ },
912
+ {
913
+ "epoch": 0.4375,
914
+ "eval_dev_acc": 0.296875,
915
+ "eval_dev_token": 5770.9375,
916
+ "eval_runtime": 175.5918,
917
+ "eval_samples_per_second": 0.091,
918
+ "eval_steps_per_second": 0.006,
919
+ "step": 112
920
+ },
921
+ {
922
+ "epoch": 0.44140625,
923
+ "grad_norm": 0.4499041384963393,
924
+ "learning_rate": 1e-05,
925
+ "loss": 0.1845,
926
+ "step": 113
927
+ },
928
+ {
929
+ "epoch": 0.4453125,
930
+ "grad_norm": 0.5818915994231291,
931
+ "learning_rate": 1e-05,
932
+ "loss": 0.2709,
933
+ "step": 114
934
+ },
935
+ {
936
+ "epoch": 0.44921875,
937
+ "grad_norm": 0.6130904000526848,
938
+ "learning_rate": 1e-05,
939
+ "loss": 0.231,
940
+ "step": 115
941
+ },
942
+ {
943
+ "epoch": 0.453125,
944
+ "grad_norm": 0.7266034880537791,
945
+ "learning_rate": 1e-05,
946
+ "loss": 0.1555,
947
+ "step": 116
948
+ },
949
+ {
950
+ "epoch": 0.45703125,
951
+ "grad_norm": 0.425032745279421,
952
+ "learning_rate": 1e-05,
953
+ "loss": 0.1733,
954
+ "step": 117
955
+ },
956
+ {
957
+ "epoch": 0.4609375,
958
+ "grad_norm": 0.41408811254876093,
959
+ "learning_rate": 1e-05,
960
+ "loss": 0.1793,
961
+ "step": 118
962
+ },
963
+ {
964
+ "epoch": 0.46484375,
965
+ "grad_norm": 0.8433491024471641,
966
+ "learning_rate": 1e-05,
967
+ "loss": 0.2335,
968
+ "step": 119
969
+ },
970
+ {
971
+ "epoch": 0.46875,
972
+ "grad_norm": 0.5585183306922875,
973
+ "learning_rate": 1e-05,
974
+ "loss": 0.2515,
975
+ "step": 120
976
+ },
977
+ {
978
+ "epoch": 0.46875,
979
+ "eval_dev_acc": 0.4724409580230713,
980
+ "eval_dev_token": 4777.55126953125,
981
+ "eval_runtime": 165.1485,
982
+ "eval_samples_per_second": 0.097,
983
+ "eval_steps_per_second": 0.006,
984
+ "step": 120
985
+ },
986
+ {
987
+ "epoch": 0.47265625,
988
+ "grad_norm": 0.9520218462259554,
989
+ "learning_rate": 1e-05,
990
+ "loss": 0.2613,
991
+ "step": 121
992
+ },
993
+ {
994
+ "epoch": 0.4765625,
995
+ "grad_norm": 0.4858585527334522,
996
+ "learning_rate": 1e-05,
997
+ "loss": 0.2379,
998
+ "step": 122
999
+ },
1000
+ {
1001
+ "epoch": 0.48046875,
1002
+ "grad_norm": 0.5772160567620949,
1003
+ "learning_rate": 1e-05,
1004
+ "loss": 0.241,
1005
+ "step": 123
1006
+ },
1007
+ {
1008
+ "epoch": 0.484375,
1009
+ "grad_norm": 0.731954162407159,
1010
+ "learning_rate": 1e-05,
1011
+ "loss": 0.2482,
1012
+ "step": 124
1013
+ },
1014
+ {
1015
+ "epoch": 0.48828125,
1016
+ "grad_norm": 0.49226621710163243,
1017
+ "learning_rate": 1e-05,
1018
+ "loss": 0.2333,
1019
+ "step": 125
1020
+ },
1021
+ {
1022
+ "epoch": 0.4921875,
1023
+ "grad_norm": 0.43779404197089106,
1024
+ "learning_rate": 1e-05,
1025
+ "loss": 0.185,
1026
+ "step": 126
1027
+ },
1028
+ {
1029
+ "epoch": 0.49609375,
1030
+ "grad_norm": 0.6856986141306837,
1031
+ "learning_rate": 1e-05,
1032
+ "loss": 0.1943,
1033
+ "step": 127
1034
+ },
1035
+ {
1036
+ "epoch": 0.5,
1037
+ "grad_norm": 0.6558122415773976,
1038
+ "learning_rate": 1e-05,
1039
+ "loss": 0.2185,
1040
+ "step": 128
1041
+ },
1042
+ {
1043
+ "epoch": 0.5,
1044
+ "eval_dev_acc": 0.4765625,
1045
+ "eval_dev_token": 4368.859375,
1046
+ "eval_runtime": 161.9718,
1047
+ "eval_samples_per_second": 0.099,
1048
+ "eval_steps_per_second": 0.006,
1049
+ "step": 128
1050
+ },
1051
+ {
1052
+ "epoch": 0.50390625,
1053
+ "grad_norm": 0.4099906022533745,
1054
+ "learning_rate": 1e-05,
1055
+ "loss": 0.2113,
1056
+ "step": 129
1057
+ },
1058
+ {
1059
+ "epoch": 0.5078125,
1060
+ "grad_norm": 0.49752415105495956,
1061
+ "learning_rate": 1e-05,
1062
+ "loss": 0.2217,
1063
+ "step": 130
1064
+ },
1065
+ {
1066
+ "epoch": 0.51171875,
1067
+ "grad_norm": 0.8912790018467623,
1068
+ "learning_rate": 1e-05,
1069
+ "loss": 0.3422,
1070
+ "step": 131
1071
+ },
1072
+ {
1073
+ "epoch": 0.515625,
1074
+ "grad_norm": 0.6764829647253893,
1075
+ "learning_rate": 1e-05,
1076
+ "loss": 0.2055,
1077
+ "step": 132
1078
+ },
1079
+ {
1080
+ "epoch": 0.51953125,
1081
+ "grad_norm": 0.8399641090693946,
1082
+ "learning_rate": 1e-05,
1083
+ "loss": 0.2087,
1084
+ "step": 133
1085
+ },
1086
+ {
1087
+ "epoch": 0.5234375,
1088
+ "grad_norm": 0.4594160953603203,
1089
+ "learning_rate": 1e-05,
1090
+ "loss": 0.2093,
1091
+ "step": 134
1092
+ },
1093
+ {
1094
+ "epoch": 0.52734375,
1095
+ "grad_norm": 0.7432138703184232,
1096
+ "learning_rate": 1e-05,
1097
+ "loss": 0.1969,
1098
+ "step": 135
1099
+ },
1100
+ {
1101
+ "epoch": 0.53125,
1102
+ "grad_norm": 0.4584467325236011,
1103
+ "learning_rate": 1e-05,
1104
+ "loss": 0.1806,
1105
+ "step": 136
1106
+ },
1107
+ {
1108
+ "epoch": 0.53125,
1109
+ "eval_dev_acc": 0.4765625,
1110
+ "eval_dev_token": 4603.53125,
1111
+ "eval_runtime": 164.3452,
1112
+ "eval_samples_per_second": 0.097,
1113
+ "eval_steps_per_second": 0.006,
1114
+ "step": 136
1115
+ },
1116
+ {
1117
+ "epoch": 0.53515625,
1118
+ "grad_norm": 0.6458588312529675,
1119
+ "learning_rate": 1e-05,
1120
+ "loss": 0.2087,
1121
+ "step": 137
1122
+ },
1123
+ {
1124
+ "epoch": 0.5390625,
1125
+ "grad_norm": 0.7370624067340756,
1126
+ "learning_rate": 1e-05,
1127
+ "loss": 0.1854,
1128
+ "step": 138
1129
+ },
1130
+ {
1131
+ "epoch": 0.54296875,
1132
+ "grad_norm": 0.7141604462138248,
1133
+ "learning_rate": 1e-05,
1134
+ "loss": 0.2535,
1135
+ "step": 139
1136
+ },
1137
+ {
1138
+ "epoch": 0.546875,
1139
+ "grad_norm": 0.8212814690178184,
1140
+ "learning_rate": 1e-05,
1141
+ "loss": 0.1668,
1142
+ "step": 140
1143
+ },
1144
+ {
1145
+ "epoch": 0.55078125,
1146
+ "grad_norm": 0.5799692948316157,
1147
+ "learning_rate": 1e-05,
1148
+ "loss": 0.2375,
1149
+ "step": 141
1150
+ },
1151
+ {
1152
+ "epoch": 0.5546875,
1153
+ "grad_norm": 0.5333639624775814,
1154
+ "learning_rate": 1e-05,
1155
+ "loss": 0.1737,
1156
+ "step": 142
1157
+ },
1158
+ {
1159
+ "epoch": 0.55859375,
1160
+ "grad_norm": 0.4076841439195106,
1161
+ "learning_rate": 1e-05,
1162
+ "loss": 0.1627,
1163
+ "step": 143
1164
+ },
1165
+ {
1166
+ "epoch": 0.5625,
1167
+ "grad_norm": 0.4118175478201596,
1168
+ "learning_rate": 1e-05,
1169
+ "loss": 0.1576,
1170
+ "step": 144
1171
+ },
1172
+ {
1173
+ "epoch": 0.5625,
1174
+ "eval_dev_acc": 0.5234375,
1175
+ "eval_dev_token": 5125.0703125,
1176
+ "eval_runtime": 168.804,
1177
+ "eval_samples_per_second": 0.095,
1178
+ "eval_steps_per_second": 0.006,
1179
+ "step": 144
1180
+ },
1181
+ {
1182
+ "epoch": 0.56640625,
1183
+ "grad_norm": 0.5988381099011506,
1184
+ "learning_rate": 1e-05,
1185
+ "loss": 0.1656,
1186
+ "step": 145
1187
+ },
1188
+ {
1189
+ "epoch": 0.5703125,
1190
+ "grad_norm": 0.9328153493065982,
1191
+ "learning_rate": 1e-05,
1192
+ "loss": 0.1788,
1193
+ "step": 146
1194
+ },
1195
+ {
1196
+ "epoch": 0.57421875,
1197
+ "grad_norm": 0.8013592126955402,
1198
+ "learning_rate": 1e-05,
1199
+ "loss": 0.2009,
1200
+ "step": 147
1201
+ },
1202
+ {
1203
+ "epoch": 0.578125,
1204
+ "grad_norm": 0.4868159061171701,
1205
+ "learning_rate": 1e-05,
1206
+ "loss": 0.217,
1207
+ "step": 148
1208
+ },
1209
+ {
1210
+ "epoch": 0.58203125,
1211
+ "grad_norm": 0.6758953539585006,
1212
+ "learning_rate": 1e-05,
1213
+ "loss": 0.2344,
1214
+ "step": 149
1215
+ },
1216
+ {
1217
+ "epoch": 0.5859375,
1218
+ "grad_norm": 0.8609458752061137,
1219
+ "learning_rate": 1e-05,
1220
+ "loss": 0.1939,
1221
+ "step": 150
1222
+ },
1223
+ {
1224
+ "epoch": 0.58984375,
1225
+ "grad_norm": 0.45913847739444186,
1226
+ "learning_rate": 1e-05,
1227
+ "loss": 0.1691,
1228
+ "step": 151
1229
+ },
1230
+ {
1231
+ "epoch": 0.59375,
1232
+ "grad_norm": 0.8064977044716175,
1233
+ "learning_rate": 1e-05,
1234
+ "loss": 0.1949,
1235
+ "step": 152
1236
+ },
1237
+ {
1238
+ "epoch": 0.59375,
1239
+ "eval_dev_acc": 0.40625,
1240
+ "eval_dev_token": 4508.484375,
1241
+ "eval_runtime": 160.3398,
1242
+ "eval_samples_per_second": 0.1,
1243
+ "eval_steps_per_second": 0.006,
1244
+ "step": 152
1245
+ },
1246
+ {
1247
+ "epoch": 0.59765625,
1248
+ "grad_norm": 0.9904042315049291,
1249
+ "learning_rate": 1e-05,
1250
+ "loss": 0.2253,
1251
+ "step": 153
1252
+ },
1253
+ {
1254
+ "epoch": 0.6015625,
1255
+ "grad_norm": 0.5524318414569037,
1256
+ "learning_rate": 1e-05,
1257
+ "loss": 0.2535,
1258
+ "step": 154
1259
+ },
1260
+ {
1261
+ "epoch": 0.60546875,
1262
+ "grad_norm": 0.418186463867415,
1263
+ "learning_rate": 1e-05,
1264
+ "loss": 0.1884,
1265
+ "step": 155
1266
+ },
1267
+ {
1268
+ "epoch": 0.609375,
1269
+ "grad_norm": 0.6311027708045368,
1270
+ "learning_rate": 1e-05,
1271
+ "loss": 0.2408,
1272
+ "step": 156
1273
+ },
1274
+ {
1275
+ "epoch": 0.61328125,
1276
+ "grad_norm": 0.4550696199781805,
1277
+ "learning_rate": 1e-05,
1278
+ "loss": 0.173,
1279
+ "step": 157
1280
+ },
1281
+ {
1282
+ "epoch": 0.6171875,
1283
+ "grad_norm": 0.4596598696608727,
1284
+ "learning_rate": 1e-05,
1285
+ "loss": 0.1592,
1286
+ "step": 158
1287
+ },
1288
+ {
1289
+ "epoch": 0.62109375,
1290
+ "grad_norm": 0.5573937890044522,
1291
+ "learning_rate": 1e-05,
1292
+ "loss": 0.1748,
1293
+ "step": 159
1294
+ },
1295
+ {
1296
+ "epoch": 0.625,
1297
+ "grad_norm": 1.0862165315332113,
1298
+ "learning_rate": 1e-05,
1299
+ "loss": 0.2369,
1300
+ "step": 160
1301
+ },
1302
+ {
1303
+ "epoch": 0.625,
1304
+ "eval_dev_acc": 0.4296875,
1305
+ "eval_dev_token": 4869.8828125,
1306
+ "eval_runtime": 167.2914,
1307
+ "eval_samples_per_second": 0.096,
1308
+ "eval_steps_per_second": 0.006,
1309
+ "step": 160
1310
+ },
1311
+ {
1312
+ "epoch": 0.62890625,
1313
+ "grad_norm": 0.46051384064237827,
1314
+ "learning_rate": 1e-05,
1315
+ "loss": 0.2086,
1316
+ "step": 161
1317
+ },
1318
+ {
1319
+ "epoch": 0.6328125,
1320
+ "grad_norm": 0.7125397532570018,
1321
+ "learning_rate": 1e-05,
1322
+ "loss": 0.2212,
1323
+ "step": 162
1324
+ },
1325
+ {
1326
+ "epoch": 0.63671875,
1327
+ "grad_norm": 0.564820498711706,
1328
+ "learning_rate": 1e-05,
1329
+ "loss": 0.3019,
1330
+ "step": 163
1331
+ },
1332
+ {
1333
+ "epoch": 0.640625,
1334
+ "grad_norm": 0.5218656690400247,
1335
+ "learning_rate": 1e-05,
1336
+ "loss": 0.1324,
1337
+ "step": 164
1338
+ },
1339
+ {
1340
+ "epoch": 0.64453125,
1341
+ "grad_norm": 0.4994022980399308,
1342
+ "learning_rate": 1e-05,
1343
+ "loss": 0.1438,
1344
+ "step": 165
1345
+ },
1346
+ {
1347
+ "epoch": 0.6484375,
1348
+ "grad_norm": 0.7016809849517179,
1349
+ "learning_rate": 1e-05,
1350
+ "loss": 0.2791,
1351
+ "step": 166
1352
+ },
1353
+ {
1354
+ "epoch": 0.65234375,
1355
+ "grad_norm": 0.597463304680723,
1356
+ "learning_rate": 1e-05,
1357
+ "loss": 0.1749,
1358
+ "step": 167
1359
+ },
1360
+ {
1361
+ "epoch": 0.65625,
1362
+ "grad_norm": 0.5536855781273838,
1363
+ "learning_rate": 1e-05,
1364
+ "loss": 0.2391,
1365
+ "step": 168
1366
+ },
1367
+ {
1368
+ "epoch": 0.65625,
1369
+ "eval_dev_acc": 0.3203125,
1370
+ "eval_dev_token": 5451.3671875,
1371
+ "eval_runtime": 172.7574,
1372
+ "eval_samples_per_second": 0.093,
1373
+ "eval_steps_per_second": 0.006,
1374
+ "step": 168
1375
+ },
1376
+ {
1377
+ "epoch": 0.66015625,
1378
+ "grad_norm": 0.9103508979108635,
1379
+ "learning_rate": 1e-05,
1380
+ "loss": 0.2613,
1381
+ "step": 169
1382
+ },
1383
+ {
1384
+ "epoch": 0.6640625,
1385
+ "grad_norm": 0.4928845564740678,
1386
+ "learning_rate": 1e-05,
1387
+ "loss": 0.215,
1388
+ "step": 170
1389
+ },
1390
+ {
1391
+ "epoch": 0.66796875,
1392
+ "grad_norm": 0.8690405638773996,
1393
+ "learning_rate": 1e-05,
1394
+ "loss": 0.2355,
1395
+ "step": 171
1396
+ },
1397
+ {
1398
+ "epoch": 0.671875,
1399
+ "grad_norm": 0.5511255682147113,
1400
+ "learning_rate": 1e-05,
1401
+ "loss": 0.2406,
1402
+ "step": 172
1403
+ },
1404
+ {
1405
+ "epoch": 0.67578125,
1406
+ "grad_norm": 0.44346107905460214,
1407
+ "learning_rate": 1e-05,
1408
+ "loss": 0.1867,
1409
+ "step": 173
1410
+ },
1411
+ {
1412
+ "epoch": 0.6796875,
1413
+ "grad_norm": 0.4019557678019079,
1414
+ "learning_rate": 1e-05,
1415
+ "loss": 0.1488,
1416
+ "step": 174
1417
+ },
1418
+ {
1419
+ "epoch": 0.68359375,
1420
+ "grad_norm": 0.4139658009208469,
1421
+ "learning_rate": 1e-05,
1422
+ "loss": 0.1666,
1423
+ "step": 175
1424
+ },
1425
+ {
1426
+ "epoch": 0.6875,
1427
+ "grad_norm": 0.45363011716779816,
1428
+ "learning_rate": 1e-05,
1429
+ "loss": 0.2006,
1430
+ "step": 176
1431
+ },
1432
+ {
1433
+ "epoch": 0.6875,
1434
+ "eval_dev_acc": 0.3385826647281647,
1435
+ "eval_dev_token": 4971.81884765625,
1436
+ "eval_runtime": 166.9967,
1437
+ "eval_samples_per_second": 0.096,
1438
+ "eval_steps_per_second": 0.006,
1439
+ "step": 176
1440
+ },
1441
+ {
1442
+ "epoch": 0.69140625,
1443
+ "grad_norm": 0.46674698673244774,
1444
+ "learning_rate": 1e-05,
1445
+ "loss": 0.1788,
1446
+ "step": 177
1447
+ },
1448
+ {
1449
+ "epoch": 0.6953125,
1450
+ "grad_norm": 0.5396579551057291,
1451
+ "learning_rate": 1e-05,
1452
+ "loss": 0.1857,
1453
+ "step": 178
1454
+ },
1455
+ {
1456
+ "epoch": 0.69921875,
1457
+ "grad_norm": 0.42472472699800484,
1458
+ "learning_rate": 1e-05,
1459
+ "loss": 0.1707,
1460
+ "step": 179
1461
+ },
1462
+ {
1463
+ "epoch": 0.703125,
1464
+ "grad_norm": 0.4208916108378261,
1465
+ "learning_rate": 1e-05,
1466
+ "loss": 0.1736,
1467
+ "step": 180
1468
+ },
1469
+ {
1470
+ "epoch": 0.70703125,
1471
+ "grad_norm": 0.5161632347165661,
1472
+ "learning_rate": 1e-05,
1473
+ "loss": 0.2074,
1474
+ "step": 181
1475
+ },
1476
+ {
1477
+ "epoch": 0.7109375,
1478
+ "grad_norm": 0.4851147968745633,
1479
+ "learning_rate": 1e-05,
1480
+ "loss": 0.2183,
1481
+ "step": 182
1482
+ },
1483
+ {
1484
+ "epoch": 0.71484375,
1485
+ "grad_norm": 0.5286494967968609,
1486
+ "learning_rate": 1e-05,
1487
+ "loss": 0.1877,
1488
+ "step": 183
1489
+ },
1490
+ {
1491
+ "epoch": 0.71875,
1492
+ "grad_norm": 0.5399316089624949,
1493
+ "learning_rate": 1e-05,
1494
+ "loss": 0.209,
1495
+ "step": 184
1496
+ },
1497
+ {
1498
+ "epoch": 0.71875,
1499
+ "eval_dev_acc": 0.3984375,
1500
+ "eval_dev_token": 4787.84375,
1501
+ "eval_runtime": 166.2574,
1502
+ "eval_samples_per_second": 0.096,
1503
+ "eval_steps_per_second": 0.006,
1504
+ "step": 184
1505
+ },
1506
+ {
1507
+ "epoch": 0.72265625,
1508
+ "grad_norm": 0.7188938790166789,
1509
+ "learning_rate": 1e-05,
1510
+ "loss": 0.2065,
1511
+ "step": 185
1512
+ },
1513
+ {
1514
+ "epoch": 0.7265625,
1515
+ "grad_norm": 0.5843767003652576,
1516
+ "learning_rate": 1e-05,
1517
+ "loss": 0.2356,
1518
+ "step": 186
1519
+ },
1520
+ {
1521
+ "epoch": 0.73046875,
1522
+ "grad_norm": 0.4904003204685076,
1523
+ "learning_rate": 1e-05,
1524
+ "loss": 0.201,
1525
+ "step": 187
1526
+ },
1527
+ {
1528
+ "epoch": 0.734375,
1529
+ "grad_norm": 0.485266158116283,
1530
+ "learning_rate": 1e-05,
1531
+ "loss": 0.1869,
1532
+ "step": 188
1533
+ },
1534
+ {
1535
+ "epoch": 0.73828125,
1536
+ "grad_norm": 0.5242977395658632,
1537
+ "learning_rate": 1e-05,
1538
+ "loss": 0.2122,
1539
+ "step": 189
1540
+ },
1541
+ {
1542
+ "epoch": 0.7421875,
1543
+ "grad_norm": 0.5417537780138298,
1544
+ "learning_rate": 1e-05,
1545
+ "loss": 0.2799,
1546
+ "step": 190
1547
+ },
1548
+ {
1549
+ "epoch": 0.74609375,
1550
+ "grad_norm": 0.48949419193338123,
1551
+ "learning_rate": 1e-05,
1552
+ "loss": 0.212,
1553
+ "step": 191
1554
+ },
1555
+ {
1556
+ "epoch": 0.75,
1557
+ "grad_norm": 0.48118963817889204,
1558
+ "learning_rate": 1e-05,
1559
+ "loss": 0.2195,
1560
+ "step": 192
1561
+ },
1562
+ {
1563
+ "epoch": 0.75,
1564
+ "eval_dev_acc": 0.453125,
1565
+ "eval_dev_token": 5056.7421875,
1566
+ "eval_runtime": 168.273,
1567
+ "eval_samples_per_second": 0.095,
1568
+ "eval_steps_per_second": 0.006,
1569
+ "step": 192
1570
+ },
1571
+ {
1572
+ "epoch": 0.75390625,
1573
+ "grad_norm": 0.6844465372064547,
1574
+ "learning_rate": 1e-05,
1575
+ "loss": 0.1645,
1576
+ "step": 193
1577
+ },
1578
+ {
1579
+ "epoch": 0.7578125,
1580
+ "grad_norm": 0.49653100043792153,
1581
+ "learning_rate": 1e-05,
1582
+ "loss": 0.2023,
1583
+ "step": 194
1584
+ },
1585
+ {
1586
+ "epoch": 0.76171875,
1587
+ "grad_norm": 0.5539027026151374,
1588
+ "learning_rate": 1e-05,
1589
+ "loss": 0.2348,
1590
+ "step": 195
1591
+ },
1592
+ {
1593
+ "epoch": 0.765625,
1594
+ "grad_norm": 0.5003270709383194,
1595
+ "learning_rate": 1e-05,
1596
+ "loss": 0.2545,
1597
+ "step": 196
1598
+ },
1599
+ {
1600
+ "epoch": 0.76953125,
1601
+ "grad_norm": 0.5666703162116131,
1602
+ "learning_rate": 1e-05,
1603
+ "loss": 0.2739,
1604
+ "step": 197
1605
+ },
1606
+ {
1607
+ "epoch": 0.7734375,
1608
+ "grad_norm": 0.5281121627729704,
1609
+ "learning_rate": 1e-05,
1610
+ "loss": 0.1927,
1611
+ "step": 198
1612
+ },
1613
+ {
1614
+ "epoch": 0.77734375,
1615
+ "grad_norm": 0.4691586351966124,
1616
+ "learning_rate": 1e-05,
1617
+ "loss": 0.2101,
1618
+ "step": 199
1619
+ },
1620
+ {
1621
+ "epoch": 0.78125,
1622
+ "grad_norm": 0.43348894899907703,
1623
+ "learning_rate": 1e-05,
1624
+ "loss": 0.1636,
1625
+ "step": 200
1626
+ },
1627
+ {
1628
+ "epoch": 0.78125,
1629
+ "eval_dev_acc": 0.4296875,
1630
+ "eval_dev_token": 5082.265625,
1631
+ "eval_runtime": 169.7777,
1632
+ "eval_samples_per_second": 0.094,
1633
+ "eval_steps_per_second": 0.006,
1634
+ "step": 200
1635
+ },
1636
+ {
1637
+ "epoch": 0.78515625,
1638
+ "grad_norm": 0.4995118305726593,
1639
+ "learning_rate": 1e-05,
1640
+ "loss": 0.2149,
1641
+ "step": 201
1642
+ },
1643
+ {
1644
+ "epoch": 0.7890625,
1645
+ "grad_norm": 0.3958721084761467,
1646
+ "learning_rate": 1e-05,
1647
+ "loss": 0.1732,
1648
+ "step": 202
1649
+ },
1650
+ {
1651
+ "epoch": 0.79296875,
1652
+ "grad_norm": 0.4883258744044862,
1653
+ "learning_rate": 1e-05,
1654
+ "loss": 0.219,
1655
+ "step": 203
1656
+ },
1657
+ {
1658
+ "epoch": 0.796875,
1659
+ "grad_norm": 0.45472746506302575,
1660
+ "learning_rate": 1e-05,
1661
+ "loss": 0.2187,
1662
+ "step": 204
1663
+ },
1664
+ {
1665
+ "epoch": 0.80078125,
1666
+ "grad_norm": 0.45006095039367805,
1667
+ "learning_rate": 1e-05,
1668
+ "loss": 0.1924,
1669
+ "step": 205
1670
+ },
1671
+ {
1672
+ "epoch": 0.8046875,
1673
+ "grad_norm": 0.4127537232406072,
1674
+ "learning_rate": 1e-05,
1675
+ "loss": 0.1736,
1676
+ "step": 206
1677
+ },
1678
+ {
1679
+ "epoch": 0.80859375,
1680
+ "grad_norm": 0.4669392415601201,
1681
+ "learning_rate": 1e-05,
1682
+ "loss": 0.1847,
1683
+ "step": 207
1684
+ },
1685
+ {
1686
+ "epoch": 0.8125,
1687
+ "grad_norm": 0.41469363114093816,
1688
+ "learning_rate": 1e-05,
1689
+ "loss": 0.1556,
1690
+ "step": 208
1691
+ },
1692
+ {
1693
+ "epoch": 0.8125,
1694
+ "eval_dev_acc": 0.4609375,
1695
+ "eval_dev_token": 4918.28125,
1696
+ "eval_runtime": 166.5675,
1697
+ "eval_samples_per_second": 0.096,
1698
+ "eval_steps_per_second": 0.006,
1699
+ "step": 208
1700
+ },
1701
+ {
1702
+ "epoch": 0.81640625,
1703
+ "grad_norm": 0.4433576280938302,
1704
+ "learning_rate": 1e-05,
1705
+ "loss": 0.1934,
1706
+ "step": 209
1707
+ },
1708
+ {
1709
+ "epoch": 0.8203125,
1710
+ "grad_norm": 0.4355305023653351,
1711
+ "learning_rate": 1e-05,
1712
+ "loss": 0.1742,
1713
+ "step": 210
1714
+ },
1715
+ {
1716
+ "epoch": 0.82421875,
1717
+ "grad_norm": 0.44938618579632195,
1718
+ "learning_rate": 1e-05,
1719
+ "loss": 0.1902,
1720
+ "step": 211
1721
+ },
1722
+ {
1723
+ "epoch": 0.828125,
1724
+ "grad_norm": 0.5351771463999816,
1725
+ "learning_rate": 1e-05,
1726
+ "loss": 0.2148,
1727
+ "step": 212
1728
+ },
1729
+ {
1730
+ "epoch": 0.83203125,
1731
+ "grad_norm": 0.5839350362138708,
1732
+ "learning_rate": 1e-05,
1733
+ "loss": 0.275,
1734
+ "step": 213
1735
+ },
1736
+ {
1737
+ "epoch": 0.8359375,
1738
+ "grad_norm": 0.6964110745693202,
1739
+ "learning_rate": 1e-05,
1740
+ "loss": 0.2179,
1741
+ "step": 214
1742
+ },
1743
+ {
1744
+ "epoch": 0.83984375,
1745
+ "grad_norm": 0.4337830660702992,
1746
+ "learning_rate": 1e-05,
1747
+ "loss": 0.2152,
1748
+ "step": 215
1749
+ },
1750
+ {
1751
+ "epoch": 0.84375,
1752
+ "grad_norm": 0.46223312750006246,
1753
+ "learning_rate": 1e-05,
1754
+ "loss": 0.2405,
1755
+ "step": 216
1756
+ },
1757
+ {
1758
+ "epoch": 0.84375,
1759
+ "eval_dev_acc": 0.3828125,
1760
+ "eval_dev_token": 5435.3046875,
1761
+ "eval_runtime": 173.8173,
1762
+ "eval_samples_per_second": 0.092,
1763
+ "eval_steps_per_second": 0.006,
1764
+ "step": 216
1765
+ },
1766
+ {
1767
+ "epoch": 0.84765625,
1768
+ "grad_norm": 0.5541820526606585,
1769
+ "learning_rate": 1e-05,
1770
+ "loss": 0.2751,
1771
+ "step": 217
1772
+ },
1773
+ {
1774
+ "epoch": 0.8515625,
1775
+ "grad_norm": 0.4662570041545537,
1776
+ "learning_rate": 1e-05,
1777
+ "loss": 0.2142,
1778
+ "step": 218
1779
+ },
1780
+ {
1781
+ "epoch": 0.85546875,
1782
+ "grad_norm": 0.7737037625157579,
1783
+ "learning_rate": 1e-05,
1784
+ "loss": 0.2397,
1785
+ "step": 219
1786
+ },
1787
+ {
1788
+ "epoch": 0.859375,
1789
+ "grad_norm": 0.5572195616624243,
1790
+ "learning_rate": 1e-05,
1791
+ "loss": 0.2421,
1792
+ "step": 220
1793
+ },
1794
+ {
1795
+ "epoch": 0.86328125,
1796
+ "grad_norm": 0.5088509372691609,
1797
+ "learning_rate": 1e-05,
1798
+ "loss": 0.1875,
1799
+ "step": 221
1800
+ },
1801
+ {
1802
+ "epoch": 0.8671875,
1803
+ "grad_norm": 0.508699458613964,
1804
+ "learning_rate": 1e-05,
1805
+ "loss": 0.1927,
1806
+ "step": 222
1807
+ },
1808
+ {
1809
+ "epoch": 0.87109375,
1810
+ "grad_norm": 0.5150091482241945,
1811
+ "learning_rate": 1e-05,
1812
+ "loss": 0.2536,
1813
+ "step": 223
1814
+ },
1815
+ {
1816
+ "epoch": 0.875,
1817
+ "grad_norm": 0.5203627078659161,
1818
+ "learning_rate": 1e-05,
1819
+ "loss": 0.2571,
1820
+ "step": 224
1821
+ },
1822
+ {
1823
+ "epoch": 0.875,
1824
+ "eval_dev_acc": 0.3515625,
1825
+ "eval_dev_token": 5227.0859375,
1826
+ "eval_runtime": 170.2355,
1827
+ "eval_samples_per_second": 0.094,
1828
+ "eval_steps_per_second": 0.006,
1829
+ "step": 224
1830
+ },
1831
+ {
1832
+ "epoch": 0.87890625,
1833
+ "grad_norm": 0.5279392216696818,
1834
+ "learning_rate": 1e-05,
1835
+ "loss": 0.2278,
1836
+ "step": 225
1837
+ },
1838
+ {
1839
+ "epoch": 0.8828125,
1840
+ "grad_norm": 0.45017131620724865,
1841
+ "learning_rate": 1e-05,
1842
+ "loss": 0.2132,
1843
+ "step": 226
1844
+ },
1845
+ {
1846
+ "epoch": 0.88671875,
1847
+ "grad_norm": 0.48915211275869575,
1848
+ "learning_rate": 1e-05,
1849
+ "loss": 0.2627,
1850
+ "step": 227
1851
+ },
1852
+ {
1853
+ "epoch": 0.890625,
1854
+ "grad_norm": 0.4606618945421734,
1855
+ "learning_rate": 1e-05,
1856
+ "loss": 0.1528,
1857
+ "step": 228
1858
+ },
1859
+ {
1860
+ "epoch": 0.89453125,
1861
+ "grad_norm": 0.5072593200666395,
1862
+ "learning_rate": 1e-05,
1863
+ "loss": 0.2148,
1864
+ "step": 229
1865
+ },
1866
+ {
1867
+ "epoch": 0.8984375,
1868
+ "grad_norm": 0.5513069869439534,
1869
+ "learning_rate": 1e-05,
1870
+ "loss": 0.2319,
1871
+ "step": 230
1872
+ },
1873
+ {
1874
+ "epoch": 0.90234375,
1875
+ "grad_norm": 0.4917083878550277,
1876
+ "learning_rate": 1e-05,
1877
+ "loss": 0.1989,
1878
+ "step": 231
1879
+ },
1880
+ {
1881
+ "epoch": 0.90625,
1882
+ "grad_norm": 0.4027028580105545,
1883
+ "learning_rate": 1e-05,
1884
+ "loss": 0.1398,
1885
+ "step": 232
1886
+ },
1887
+ {
1888
+ "epoch": 0.90625,
1889
+ "eval_dev_acc": 0.3779527544975281,
1890
+ "eval_dev_token": 5651.6455078125,
1891
+ "eval_runtime": 175.5543,
1892
+ "eval_samples_per_second": 0.091,
1893
+ "eval_steps_per_second": 0.006,
1894
+ "step": 232
1895
+ },
1896
+ {
1897
+ "epoch": 0.91015625,
1898
+ "grad_norm": 0.4098440727615931,
1899
+ "learning_rate": 1e-05,
1900
+ "loss": 0.1481,
1901
+ "step": 233
1902
+ },
1903
+ {
1904
+ "epoch": 0.9140625,
1905
+ "grad_norm": 0.4379253949500134,
1906
+ "learning_rate": 1e-05,
1907
+ "loss": 0.172,
1908
+ "step": 234
1909
+ },
1910
+ {
1911
+ "epoch": 0.91796875,
1912
+ "grad_norm": 0.6161974608496972,
1913
+ "learning_rate": 1e-05,
1914
+ "loss": 0.2234,
1915
+ "step": 235
1916
+ },
1917
+ {
1918
+ "epoch": 0.921875,
1919
+ "grad_norm": 0.6431694552333217,
1920
+ "learning_rate": 1e-05,
1921
+ "loss": 0.2928,
1922
+ "step": 236
1923
+ },
1924
+ {
1925
+ "epoch": 0.92578125,
1926
+ "grad_norm": 0.7524837454023333,
1927
+ "learning_rate": 1e-05,
1928
+ "loss": 0.3518,
1929
+ "step": 237
1930
+ },
1931
+ {
1932
+ "epoch": 0.9296875,
1933
+ "grad_norm": 0.5137794157548315,
1934
+ "learning_rate": 1e-05,
1935
+ "loss": 0.2371,
1936
+ "step": 238
1937
+ },
1938
+ {
1939
+ "epoch": 0.93359375,
1940
+ "grad_norm": 0.42726761741926383,
1941
+ "learning_rate": 1e-05,
1942
+ "loss": 0.1349,
1943
+ "step": 239
1944
+ },
1945
+ {
1946
+ "epoch": 0.9375,
1947
+ "grad_norm": 0.50721507122848,
1948
+ "learning_rate": 1e-05,
1949
+ "loss": 0.147,
1950
+ "step": 240
1951
+ },
1952
+ {
1953
+ "epoch": 0.9375,
1954
+ "eval_dev_acc": 0.4375,
1955
+ "eval_dev_token": 5554.34375,
1956
+ "eval_runtime": 173.4206,
1957
+ "eval_samples_per_second": 0.092,
1958
+ "eval_steps_per_second": 0.006,
1959
+ "step": 240
1960
+ },
1961
+ {
1962
+ "epoch": 0.94140625,
1963
+ "grad_norm": 0.5085504060972834,
1964
+ "learning_rate": 1e-05,
1965
+ "loss": 0.2115,
1966
+ "step": 241
1967
+ },
1968
+ {
1969
+ "epoch": 0.9453125,
1970
+ "grad_norm": 0.5245333395138617,
1971
+ "learning_rate": 1e-05,
1972
+ "loss": 0.2203,
1973
+ "step": 242
1974
+ },
1975
+ {
1976
+ "epoch": 0.94921875,
1977
+ "grad_norm": 0.5149241747645703,
1978
+ "learning_rate": 1e-05,
1979
+ "loss": 0.1935,
1980
+ "step": 243
1981
+ },
1982
+ {
1983
+ "epoch": 0.953125,
1984
+ "grad_norm": 0.45199967311107936,
1985
+ "learning_rate": 1e-05,
1986
+ "loss": 0.1875,
1987
+ "step": 244
1988
+ },
1989
+ {
1990
+ "epoch": 0.95703125,
1991
+ "grad_norm": 0.6017279864923942,
1992
+ "learning_rate": 1e-05,
1993
+ "loss": 0.1964,
1994
+ "step": 245
1995
+ },
1996
+ {
1997
+ "epoch": 0.9609375,
1998
+ "grad_norm": 0.541548647166723,
1999
+ "learning_rate": 1e-05,
2000
+ "loss": 0.2029,
2001
+ "step": 246
2002
+ },
2003
+ {
2004
+ "epoch": 0.96484375,
2005
+ "grad_norm": 0.7095706252744872,
2006
+ "learning_rate": 1e-05,
2007
+ "loss": 0.1824,
2008
+ "step": 247
2009
+ },
2010
+ {
2011
+ "epoch": 0.96875,
2012
+ "grad_norm": 0.6630534512223186,
2013
+ "learning_rate": 1e-05,
2014
+ "loss": 0.2346,
2015
+ "step": 248
2016
+ },
2017
+ {
2018
+ "epoch": 0.96875,
2019
+ "eval_dev_acc": 0.5234375,
2020
+ "eval_dev_token": 5464.203125,
2021
+ "eval_runtime": 173.0858,
2022
+ "eval_samples_per_second": 0.092,
2023
+ "eval_steps_per_second": 0.006,
2024
+ "step": 248
2025
+ },
2026
+ {
2027
+ "epoch": 0.97265625,
2028
+ "grad_norm": 0.7470938668923351,
2029
+ "learning_rate": 1e-05,
2030
+ "loss": 0.3028,
2031
+ "step": 249
2032
+ },
2033
+ {
2034
+ "epoch": 0.9765625,
2035
+ "grad_norm": 0.534162369114681,
2036
+ "learning_rate": 1e-05,
2037
+ "loss": 0.243,
2038
+ "step": 250
2039
+ },
2040
+ {
2041
+ "epoch": 0.98046875,
2042
+ "grad_norm": 0.5240149993617814,
2043
+ "learning_rate": 1e-05,
2044
+ "loss": 0.2475,
2045
+ "step": 251
2046
+ },
2047
+ {
2048
+ "epoch": 0.984375,
2049
+ "grad_norm": 0.48058164633897993,
2050
+ "learning_rate": 1e-05,
2051
+ "loss": 0.2234,
2052
+ "step": 252
2053
+ },
2054
+ {
2055
+ "epoch": 0.98828125,
2056
+ "grad_norm": 0.5427424821749397,
2057
+ "learning_rate": 1e-05,
2058
+ "loss": 0.2338,
2059
+ "step": 253
2060
+ },
2061
+ {
2062
+ "epoch": 0.9921875,
2063
+ "grad_norm": 0.5309304323745797,
2064
+ "learning_rate": 1e-05,
2065
+ "loss": 0.2751,
2066
+ "step": 254
2067
+ },
2068
+ {
2069
+ "epoch": 0.99609375,
2070
+ "grad_norm": 0.4961154954055658,
2071
+ "learning_rate": 1e-05,
2072
+ "loss": 0.2329,
2073
+ "step": 255
2074
+ },
2075
+ {
2076
+ "epoch": 1.0,
2077
+ "grad_norm": 0.519835488758917,
2078
+ "learning_rate": 1e-05,
2079
+ "loss": 0.2182,
2080
+ "step": 256
2081
+ },
2082
+ {
2083
+ "epoch": 1.0,
2084
+ "eval_dev_acc": 0.4453125,
2085
+ "eval_dev_token": 5674.0546875,
2086
+ "eval_runtime": 175.8662,
2087
+ "eval_samples_per_second": 0.091,
2088
+ "eval_steps_per_second": 0.006,
2089
+ "step": 256
2090
+ }
2091
+ ],
2092
+ "logging_steps": 1.0,
2093
+ "max_steps": 256,
2094
+ "num_input_tokens_seen": 0,
2095
+ "num_train_epochs": 9223372036854775807,
2096
+ "save_steps": 64,
2097
+ "stateful_callbacks": {
2098
+ "TrainerControl": {
2099
+ "args": {
2100
+ "should_epoch_stop": false,
2101
+ "should_evaluate": false,
2102
+ "should_log": false,
2103
+ "should_save": true,
2104
+ "should_training_stop": true
2105
+ },
2106
+ "attributes": {}
2107
+ }
2108
+ },
2109
+ "total_flos": 31380919492608.0,
2110
+ "train_batch_size": 8,
2111
+ "trial_name": null,
2112
+ "trial_params": null
2113
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:293697045c82976ebdb828b71e8c654446c5ec0cc96c6a95e6cd39036cbaa551
3
+ size 8376
zero_to_fp32.py ADDED
@@ -0,0 +1,674 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ #!/usr/bin/env python
2
+
3
+ # Copyright (c) Microsoft Corporation.
4
+ # SPDX-License-Identifier: Apache-2.0
5
+
6
+ # DeepSpeed Team
7
+
8
+ # This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
9
+ # copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
10
+ # the future. Once extracted, the weights don't require DeepSpeed and can be used in any
11
+ # application.
12
+ #
13
+ # example:
14
+ # python zero_to_fp32.py . output_dir/
15
+ # or
16
+ # python zero_to_fp32.py . output_dir/ --safe_serialization
17
+
18
+ import argparse
19
+ import torch
20
+ import glob
21
+ import math
22
+ import os
23
+ import re
24
+ import json
25
+ from tqdm import tqdm
26
+ from collections import OrderedDict
27
+ from dataclasses import dataclass
28
+
29
+ # while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
30
+ # DeepSpeed data structures it has to be available in the current python environment.
31
+ from deepspeed.utils import logger
32
+ from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
33
+ FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
34
+ FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
35
+
36
+
37
+ @dataclass
38
+ class zero_model_state:
39
+ buffers: dict()
40
+ param_shapes: dict()
41
+ shared_params: list
42
+ ds_version: int
43
+ frozen_param_shapes: dict()
44
+ frozen_param_fragments: dict()
45
+
46
+
47
+ debug = 0
48
+
49
+ # load to cpu
50
+ device = torch.device('cpu')
51
+
52
+
53
+ def atoi(text):
54
+ return int(text) if text.isdigit() else text
55
+
56
+
57
+ def natural_keys(text):
58
+ '''
59
+ alist.sort(key=natural_keys) sorts in human order
60
+ http://nedbatchelder.com/blog/200712/human_sorting.html
61
+ (See Toothy's implementation in the comments)
62
+ '''
63
+ return [atoi(c) for c in re.split(r'(\d+)', text)]
64
+
65
+
66
+ def get_model_state_file(checkpoint_dir, zero_stage):
67
+ if not os.path.isdir(checkpoint_dir):
68
+ raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
69
+
70
+ # there should be only one file
71
+ if zero_stage <= 2:
72
+ file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
73
+ elif zero_stage == 3:
74
+ file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
75
+
76
+ if not os.path.exists(file):
77
+ raise FileNotFoundError(f"can't find model states file at '{file}'")
78
+
79
+ return file
80
+
81
+
82
+ def get_checkpoint_files(checkpoint_dir, glob_pattern):
83
+ # XXX: need to test that this simple glob rule works for multi-node setup too
84
+ ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
85
+
86
+ if len(ckpt_files) == 0:
87
+ raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
88
+
89
+ return ckpt_files
90
+
91
+
92
+ def get_optim_files(checkpoint_dir):
93
+ return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
94
+
95
+
96
+ def get_model_state_files(checkpoint_dir):
97
+ return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
98
+
99
+
100
+ def parse_model_states(files):
101
+ zero_model_states = []
102
+ for file in files:
103
+ state_dict = torch.load(file, map_location=device)
104
+
105
+ if BUFFER_NAMES not in state_dict:
106
+ raise ValueError(f"{file} is not a model state checkpoint")
107
+ buffer_names = state_dict[BUFFER_NAMES]
108
+ if debug:
109
+ print("Found buffers:", buffer_names)
110
+
111
+ # recover just the buffers while restoring them to fp32 if they were saved in fp16
112
+ buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
113
+ param_shapes = state_dict[PARAM_SHAPES]
114
+
115
+ # collect parameters that are included in param_shapes
116
+ param_names = []
117
+ for s in param_shapes:
118
+ for name in s.keys():
119
+ param_names.append(name)
120
+
121
+ # update with frozen parameters
122
+ frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
123
+ if frozen_param_shapes is not None:
124
+ if debug:
125
+ print(f"Found frozen_param_shapes: {frozen_param_shapes}")
126
+ param_names += list(frozen_param_shapes.keys())
127
+
128
+ # handle shared params
129
+ shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
130
+
131
+ ds_version = state_dict.get(DS_VERSION, None)
132
+
133
+ frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
134
+
135
+ z_model_state = zero_model_state(buffers=buffers,
136
+ param_shapes=param_shapes,
137
+ shared_params=shared_params,
138
+ ds_version=ds_version,
139
+ frozen_param_shapes=frozen_param_shapes,
140
+ frozen_param_fragments=frozen_param_fragments)
141
+ zero_model_states.append(z_model_state)
142
+
143
+ return zero_model_states
144
+
145
+
146
+ def parse_optim_states(files, ds_checkpoint_dir):
147
+ total_files = len(files)
148
+ state_dicts = []
149
+ for f in files:
150
+ state_dict = torch.load(f, map_location=device)
151
+ # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
152
+ # and also handle the case where it was already removed by another helper script
153
+ state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
154
+ state_dicts.append(state_dict)
155
+
156
+ if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
157
+ raise ValueError(f"{files[0]} is not a zero checkpoint")
158
+ zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
159
+ world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
160
+
161
+ # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
162
+ # parameters can be different from data parallelism for non-expert parameters. So we can just
163
+ # use the max of the partition_count to get the dp world_size.
164
+
165
+ if type(world_size) is list:
166
+ world_size = max(world_size)
167
+
168
+ if world_size != total_files:
169
+ raise ValueError(
170
+ f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
171
+ "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
172
+ )
173
+
174
+ # the groups are named differently in each stage
175
+ if zero_stage <= 2:
176
+ fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
177
+ elif zero_stage == 3:
178
+ fp32_groups_key = FP32_FLAT_GROUPS
179
+ else:
180
+ raise ValueError(f"unknown zero stage {zero_stage}")
181
+
182
+ if zero_stage <= 2:
183
+ fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
184
+ elif zero_stage == 3:
185
+ # if there is more than one param group, there will be multiple flattened tensors - one
186
+ # flattened tensor per group - for simplicity merge them into a single tensor
187
+ #
188
+ # XXX: could make the script more memory efficient for when there are multiple groups - it
189
+ # will require matching the sub-lists of param_shapes for each param group flattened tensor
190
+
191
+ fp32_flat_groups = [
192
+ torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts))
193
+ ]
194
+
195
+ return zero_stage, world_size, fp32_flat_groups
196
+
197
+
198
+ def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
199
+ """
200
+ Returns fp32 state_dict reconstructed from ds checkpoint
201
+
202
+ Args:
203
+ - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
204
+
205
+ """
206
+ print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
207
+
208
+ optim_files = get_optim_files(ds_checkpoint_dir)
209
+ zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
210
+ print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
211
+
212
+ model_files = get_model_state_files(ds_checkpoint_dir)
213
+
214
+ zero_model_states = parse_model_states(model_files)
215
+ print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
216
+
217
+ if zero_stage <= 2:
218
+ return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
219
+ exclude_frozen_parameters)
220
+ elif zero_stage == 3:
221
+ return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
222
+ exclude_frozen_parameters)
223
+
224
+
225
+ def _zero2_merge_frozen_params(state_dict, zero_model_states):
226
+ if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
227
+ return
228
+
229
+ frozen_param_shapes = zero_model_states[0].frozen_param_shapes
230
+ frozen_param_fragments = zero_model_states[0].frozen_param_fragments
231
+
232
+ if debug:
233
+ num_elem = sum(s.numel() for s in frozen_param_shapes.values())
234
+ print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
235
+
236
+ wanted_params = len(frozen_param_shapes)
237
+ wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
238
+ avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
239
+ print(f'Frozen params: Have {avail_numel} numels to process.')
240
+ print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
241
+
242
+ total_params = 0
243
+ total_numel = 0
244
+ for name, shape in frozen_param_shapes.items():
245
+ total_params += 1
246
+ unpartitioned_numel = shape.numel()
247
+ total_numel += unpartitioned_numel
248
+
249
+ state_dict[name] = frozen_param_fragments[name]
250
+
251
+ if debug:
252
+ print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
253
+
254
+ print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
255
+
256
+
257
+ def _has_callable(obj, fn):
258
+ attr = getattr(obj, fn, None)
259
+ return callable(attr)
260
+
261
+
262
+ def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
263
+ param_shapes = zero_model_states[0].param_shapes
264
+
265
+ # Reconstruction protocol:
266
+ #
267
+ # XXX: document this
268
+
269
+ if debug:
270
+ for i in range(world_size):
271
+ for j in range(len(fp32_flat_groups[0])):
272
+ print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
273
+
274
+ # XXX: memory usage doubles here (zero2)
275
+ num_param_groups = len(fp32_flat_groups[0])
276
+ merged_single_partition_of_fp32_groups = []
277
+ for i in range(num_param_groups):
278
+ merged_partitions = [sd[i] for sd in fp32_flat_groups]
279
+ full_single_fp32_vector = torch.cat(merged_partitions, 0)
280
+ merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
281
+ avail_numel = sum(
282
+ [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
283
+
284
+ if debug:
285
+ wanted_params = sum([len(shapes) for shapes in param_shapes])
286
+ wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
287
+ # not asserting if there is a mismatch due to possible padding
288
+ print(f"Have {avail_numel} numels to process.")
289
+ print(f"Need {wanted_numel} numels in {wanted_params} params.")
290
+
291
+ # params
292
+ # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
293
+ # out-of-core computing solution
294
+ total_numel = 0
295
+ total_params = 0
296
+ for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
297
+ offset = 0
298
+ avail_numel = full_single_fp32_vector.numel()
299
+ for name, shape in shapes.items():
300
+
301
+ unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
302
+ total_numel += unpartitioned_numel
303
+ total_params += 1
304
+
305
+ if debug:
306
+ print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
307
+ state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
308
+ offset += unpartitioned_numel
309
+
310
+ # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
311
+ # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
312
+ # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
313
+ # live optimizer object, so we are checking that the numbers are within the right range
314
+ align_to = 2 * world_size
315
+
316
+ def zero2_align(x):
317
+ return align_to * math.ceil(x / align_to)
318
+
319
+ if debug:
320
+ print(f"original offset={offset}, avail_numel={avail_numel}")
321
+
322
+ offset = zero2_align(offset)
323
+ avail_numel = zero2_align(avail_numel)
324
+
325
+ if debug:
326
+ print(f"aligned offset={offset}, avail_numel={avail_numel}")
327
+
328
+ # Sanity check
329
+ if offset != avail_numel:
330
+ raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
331
+
332
+ print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
333
+
334
+
335
+ def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
336
+ exclude_frozen_parameters):
337
+ state_dict = OrderedDict()
338
+
339
+ # buffers
340
+ buffers = zero_model_states[0].buffers
341
+ state_dict.update(buffers)
342
+ if debug:
343
+ print(f"added {len(buffers)} buffers")
344
+
345
+ if not exclude_frozen_parameters:
346
+ _zero2_merge_frozen_params(state_dict, zero_model_states)
347
+
348
+ _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
349
+
350
+ # recover shared parameters
351
+ for pair in zero_model_states[0].shared_params:
352
+ if pair[1] in state_dict:
353
+ state_dict[pair[0]] = state_dict[pair[1]]
354
+
355
+ return state_dict
356
+
357
+
358
+ def zero3_partitioned_param_info(unpartitioned_numel, world_size):
359
+ remainder = unpartitioned_numel % world_size
360
+ padding_numel = (world_size - remainder) if remainder else 0
361
+ partitioned_numel = math.ceil(unpartitioned_numel / world_size)
362
+ return partitioned_numel, padding_numel
363
+
364
+
365
+ def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
366
+ if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
367
+ return
368
+
369
+ if debug:
370
+ for i in range(world_size):
371
+ num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
372
+ print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
373
+
374
+ frozen_param_shapes = zero_model_states[0].frozen_param_shapes
375
+ wanted_params = len(frozen_param_shapes)
376
+ wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
377
+ avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
378
+ print(f'Frozen params: Have {avail_numel} numels to process.')
379
+ print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
380
+
381
+ total_params = 0
382
+ total_numel = 0
383
+ for name, shape in zero_model_states[0].frozen_param_shapes.items():
384
+ total_params += 1
385
+ unpartitioned_numel = shape.numel()
386
+ total_numel += unpartitioned_numel
387
+
388
+ param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
389
+ state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
390
+
391
+ partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
392
+
393
+ if debug:
394
+ print(
395
+ f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
396
+ )
397
+
398
+ print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
399
+
400
+
401
+ def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
402
+ param_shapes = zero_model_states[0].param_shapes
403
+ avail_numel = fp32_flat_groups[0].numel() * world_size
404
+ # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
405
+ # param, re-consolidating each param, while dealing with padding if any
406
+
407
+ # merge list of dicts, preserving order
408
+ param_shapes = {k: v for d in param_shapes for k, v in d.items()}
409
+
410
+ if debug:
411
+ for i in range(world_size):
412
+ print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
413
+
414
+ wanted_params = len(param_shapes)
415
+ wanted_numel = sum(shape.numel() for shape in param_shapes.values())
416
+ # not asserting if there is a mismatch due to possible padding
417
+ avail_numel = fp32_flat_groups[0].numel() * world_size
418
+ print(f"Trainable params: Have {avail_numel} numels to process.")
419
+ print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
420
+
421
+ # params
422
+ # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
423
+ # out-of-core computing solution
424
+ offset = 0
425
+ total_numel = 0
426
+ total_params = 0
427
+ for name, shape in tqdm(param_shapes.items(), desc='Gathering Sharded Weights'):
428
+ unpartitioned_numel = shape.numel()
429
+ total_numel += unpartitioned_numel
430
+ total_params += 1
431
+ partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
432
+
433
+ if debug:
434
+ print(
435
+ f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
436
+ )
437
+
438
+ # XXX: memory usage doubles here
439
+ state_dict[name] = torch.cat(
440
+ tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)),
441
+ 0).narrow(0, 0, unpartitioned_numel).view(shape)
442
+ offset += partitioned_numel
443
+
444
+ offset *= world_size
445
+
446
+ # Sanity check
447
+ if offset != avail_numel:
448
+ raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
449
+
450
+ print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
451
+
452
+
453
+ def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
454
+ exclude_frozen_parameters):
455
+ state_dict = OrderedDict()
456
+
457
+ # buffers
458
+ buffers = zero_model_states[0].buffers
459
+ state_dict.update(buffers)
460
+ if debug:
461
+ print(f"added {len(buffers)} buffers")
462
+
463
+ if not exclude_frozen_parameters:
464
+ _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
465
+
466
+ _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
467
+
468
+ # recover shared parameters
469
+ for pair in zero_model_states[0].shared_params:
470
+ if pair[1] in state_dict:
471
+ state_dict[pair[0]] = state_dict[pair[1]]
472
+
473
+ return state_dict
474
+
475
+
476
+ def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False):
477
+ """
478
+ Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
479
+ ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
480
+ via a model hub.
481
+
482
+ Args:
483
+ - ``checkpoint_dir``: path to the desired checkpoint folder
484
+ - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
485
+ - ``exclude_frozen_parameters``: exclude frozen parameters
486
+
487
+ Returns:
488
+ - pytorch ``state_dict``
489
+
490
+ Note: this approach may not work if your application doesn't have sufficient free CPU memory and
491
+ you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
492
+ the checkpoint.
493
+
494
+ A typical usage might be ::
495
+
496
+ from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
497
+ # do the training and checkpoint saving
498
+ state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
499
+ model = model.cpu() # move to cpu
500
+ model.load_state_dict(state_dict)
501
+ # submit to model hub or save the model to share with others
502
+
503
+ In this example the ``model`` will no longer be usable in the deepspeed context of the same
504
+ application. i.e. you will need to re-initialize the deepspeed engine, since
505
+ ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
506
+
507
+ If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
508
+
509
+ """
510
+ if tag is None:
511
+ latest_path = os.path.join(checkpoint_dir, 'latest')
512
+ if os.path.isfile(latest_path):
513
+ with open(latest_path, 'r') as fd:
514
+ tag = fd.read().strip()
515
+ else:
516
+ raise ValueError(f"Unable to find 'latest' file at {latest_path}")
517
+
518
+ ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
519
+
520
+ if not os.path.isdir(ds_checkpoint_dir):
521
+ raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
522
+
523
+ return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
524
+
525
+
526
+ def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
527
+ output_dir,
528
+ max_shard_size="5GB",
529
+ safe_serialization=False,
530
+ tag=None,
531
+ exclude_frozen_parameters=False):
532
+ """
533
+ Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
534
+ loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
535
+
536
+ Args:
537
+ - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
538
+ - ``output_dir``: directory to the pytorch fp32 state_dict output files
539
+ - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
540
+ - ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
541
+ - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
542
+ - ``exclude_frozen_parameters``: exclude frozen parameters
543
+ """
544
+ # Dependency pre-check
545
+ if safe_serialization:
546
+ try:
547
+ from safetensors.torch import save_file
548
+ except ImportError:
549
+ print('If you want to use `safe_serialization`, please `pip install safetensors`')
550
+ raise
551
+ if max_shard_size is not None:
552
+ try:
553
+ from huggingface_hub import split_torch_state_dict_into_shards
554
+ except ImportError:
555
+ print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
556
+ raise
557
+
558
+ # Convert zero checkpoint to state_dict
559
+ state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters)
560
+
561
+ # Shard the model if it is too big.
562
+ weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
563
+ if max_shard_size is not None:
564
+ filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
565
+ state_dict_split = split_torch_state_dict_into_shards(state_dict,
566
+ filename_pattern=filename_pattern,
567
+ max_shard_size=max_shard_size)
568
+ else:
569
+ from collections import namedtuple
570
+ StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
571
+ state_dict_split = StateDictSplit(is_sharded=False,
572
+ filename_to_tensors={weights_name: list(state_dict.keys())})
573
+
574
+ # Save the model
575
+ filename_to_tensors = state_dict_split.filename_to_tensors.items()
576
+ for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
577
+ shard = {tensor: state_dict[tensor].contiguous() for tensor in tensors}
578
+ output_path = os.path.join(output_dir, shard_file)
579
+ if safe_serialization:
580
+ save_file(shard, output_path, metadata={"format": "pt"})
581
+ else:
582
+ torch.save(shard, output_path)
583
+
584
+ # Save index if sharded
585
+ if state_dict_split.is_sharded:
586
+ index = {
587
+ "metadata": state_dict_split.metadata,
588
+ "weight_map": state_dict_split.tensor_to_filename,
589
+ }
590
+ save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
591
+ save_index_file = os.path.join(output_dir, save_index_file)
592
+ with open(save_index_file, "w", encoding="utf-8") as f:
593
+ content = json.dumps(index, indent=2, sort_keys=True) + "\n"
594
+ f.write(content)
595
+
596
+
597
+ def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
598
+ """
599
+ 1. Put the provided model to cpu
600
+ 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
601
+ 3. Load it into the provided model
602
+
603
+ Args:
604
+ - ``model``: the model object to update
605
+ - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
606
+ - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
607
+
608
+ Returns:
609
+ - ``model`: modified model
610
+
611
+ Make sure you have plenty of CPU memory available before you call this function. If you don't
612
+ have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
613
+ conveniently placed for you in the checkpoint folder.
614
+
615
+ A typical usage might be ::
616
+
617
+ from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
618
+ model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
619
+ # submit to model hub or save the model to share with others
620
+
621
+ Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
622
+ of the same application. i.e. you will need to re-initialize the deepspeed engine, since
623
+ ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
624
+
625
+ """
626
+ logger.info(f"Extracting fp32 weights")
627
+ state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
628
+
629
+ logger.info(f"Overwriting model with fp32 weights")
630
+ model = model.cpu()
631
+ model.load_state_dict(state_dict, strict=False)
632
+
633
+ return model
634
+
635
+
636
+ if __name__ == "__main__":
637
+ parser = argparse.ArgumentParser()
638
+ parser.add_argument("checkpoint_dir",
639
+ type=str,
640
+ help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
641
+ parser.add_argument("output_dir",
642
+ type=str,
643
+ help="directory to the pytorch fp32 state_dict output files"
644
+ "(e.g. path/checkpoint-12-output/)")
645
+ parser.add_argument(
646
+ "--max_shard_size",
647
+ type=str,
648
+ default="5GB",
649
+ help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
650
+ "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
651
+ "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
652
+ "without CPU OOM issues.")
653
+ parser.add_argument(
654
+ "--safe_serialization",
655
+ default=False,
656
+ action='store_true',
657
+ help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
658
+ parser.add_argument("-t",
659
+ "--tag",
660
+ type=str,
661
+ default=None,
662
+ help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
663
+ parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
664
+ parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
665
+ args = parser.parse_args()
666
+
667
+ debug = args.debug
668
+
669
+ convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
670
+ args.output_dir,
671
+ max_shard_size=args.max_shard_size,
672
+ safe_serialization=args.safe_serialization,
673
+ tag=args.tag,
674
+ exclude_frozen_parameters=args.exclude_frozen_parameters)