Hugging Face's logo

noumenon-labs
/
Sniff-0.6B

Safetensors
English
qwen3
Model card Files Community
marcuscedricridia commited on
Commit
606a7c3
·
verified ·
1 Parent(s): 23f21cf

Upload folder using huggingface_hub

Browse files
Files changed (16) hide show
  1. .gitattributes +1 -0
  2. added_tokens.json +28 -0
  3. config.json +32 -0
  4. generation_config.json +8 -0
  5. merges.txt +0 -0
  6. model.safetensors +3 -0
  7. optimizer.pt +3 -0
  8. rng_state.pth +3 -0
  9. scaler.pt +3 -0
  10. scheduler.pt +3 -0
  11. special_tokens_map.json +31 -0
  12. tokenizer.json +3 -0
  13. tokenizer_config.json +240 -0
  14. trainer_state.json +1154 -0
  15. training_args.bin +3 -0
  16. vocab.json +0 -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
added_tokens.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</think>": 151668,
3
+ "</tool_call>": 151658,
4
+ "</tool_response>": 151666,
5
+ "<think>": 151667,
6
+ "<tool_call>": 151657,
7
+ "<tool_response>": 151665,
8
+ "<|box_end|>": 151649,
9
+ "<|box_start|>": 151648,
10
+ "<|endoftext|>": 151643,
11
+ "<|file_sep|>": 151664,
12
+ "<|fim_middle|>": 151660,
13
+ "<|fim_pad|>": 151662,
14
+ "<|fim_prefix|>": 151659,
15
+ "<|fim_suffix|>": 151661,
16
+ "<|im_end|>": 151645,
17
+ "<|im_start|>": 151644,
18
+ "<|image_pad|>": 151655,
19
+ "<|object_ref_end|>": 151647,
20
+ "<|object_ref_start|>": 151646,
21
+ "<|quad_end|>": 151651,
22
+ "<|quad_start|>": 151650,
23
+ "<|repo_name|>": 151663,
24
+ "<|video_pad|>": 151656,
25
+ "<|vision_end|>": 151653,
26
+ "<|vision_pad|>": 151654,
27
+ "<|vision_start|>": 151652
28
+ }
config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "eos_token_id": 151643,
8
+ "head_dim": 128,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 1024,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 3072,
13
+ "max_position_embeddings": 32768,
14
+ "max_window_layers": 28,
15
+ "model_type": "qwen3",
16
+ "num_attention_heads": 16,
17
+ "num_hidden_layers": 28,
18
+ "num_key_value_heads": 8,
19
+ "pad_token_id": 151654,
20
+ "rms_norm_eps": 1e-06,
21
+ "rope_scaling": null,
22
+ "rope_theta": 1000000,
23
+ "sliding_window": null,
24
+ "tie_word_embeddings": true,
25
+ "torch_dtype": "float32",
26
+ "transformers_version": "4.52.4",
27
+ "unsloth_fixed": true,
28
+ "unsloth_version": "2025.6.4",
29
+ "use_cache": true,
30
+ "use_sliding_window": false,
31
+ "vocab_size": 151936
32
+ }
generation_config.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "eos_token_id": 151643,
4
+ "max_length": 32768,
5
+ "max_new_tokens": 2048,
6
+ "pad_token_id": 151654,
7
+ "transformers_version": "4.52.4"
8
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cd091c135a6b13254575a6c3262bd64732084f2b6f336944b7ee1f21dabf6b84
3
+ size 2384234968
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:521ed3f854964a46335a37e8a238841593fff3aa7c38f1e313efa0d7729b53d6
3
+ size 2140048520
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ba77c4358d5913436545fc6995706402cd54ccf015646708e622eca7f93ed87
3
+ size 14244
scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba07ae31ec5c29c68b2978c85c45cd4a550d5e8a440574a3d9d1a214f3baa591
3
+ size 988
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:47409458100b5a48f3a1b8bf3cbcba48c4596625e1b289ffb5ad19fc7a770ca1
3
+ size 1064
special_tokens_map.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|object_ref_start|>",
6
+ "<|object_ref_end|>",
7
+ "<|box_start|>",
8
+ "<|box_end|>",
9
+ "<|quad_start|>",
10
+ "<|quad_end|>",
11
+ "<|vision_start|>",
12
+ "<|vision_end|>",
13
+ "<|vision_pad|>",
14
+ "<|image_pad|>",
15
+ "<|video_pad|>"
16
+ ],
17
+ "eos_token": {
18
+ "content": "<|endoftext|>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "pad_token": {
25
+ "content": "<|vision_pad|>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
3
+ size 11422654
tokenizer_config.json ADDED
@@ -0,0 +1,240 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ },
181
+ "151665": {
182
+ "content": "<tool_response>",
183
+ "lstrip": false,
184
+ "normalized": false,
185
+ "rstrip": false,
186
+ "single_word": false,
187
+ "special": false
188
+ },
189
+ "151666": {
190
+ "content": "</tool_response>",
191
+ "lstrip": false,
192
+ "normalized": false,
193
+ "rstrip": false,
194
+ "single_word": false,
195
+ "special": false
196
+ },
197
+ "151667": {
198
+ "content": "<think>",
199
+ "lstrip": false,
200
+ "normalized": false,
201
+ "rstrip": false,
202
+ "single_word": false,
203
+ "special": false
204
+ },
205
+ "151668": {
206
+ "content": "</think>",
207
+ "lstrip": false,
208
+ "normalized": false,
209
+ "rstrip": false,
210
+ "single_word": false,
211
+ "special": false
212
+ }
213
+ },
214
+ "additional_special_tokens": [
215
+ "<|im_start|>",
216
+ "<|im_end|>",
217
+ "<|object_ref_start|>",
218
+ "<|object_ref_end|>",
219
+ "<|box_start|>",
220
+ "<|box_end|>",
221
+ "<|quad_start|>",
222
+ "<|quad_end|>",
223
+ "<|vision_start|>",
224
+ "<|vision_end|>",
225
+ "<|vision_pad|>",
226
+ "<|image_pad|>",
227
+ "<|video_pad|>"
228
+ ],
229
+ "bos_token": null,
230
+ "clean_up_tokenization_spaces": false,
231
+ "eos_token": "<|endoftext|>",
232
+ "errors": "replace",
233
+ "extra_special_tokens": {},
234
+ "model_max_length": 32768,
235
+ "pad_token": "<|vision_pad|>",
236
+ "padding_side": "right",
237
+ "split_special_tokens": false,
238
+ "tokenizer_class": "Qwen2Tokenizer",
239
+ "unk_token": null
240
+ }
trainer_state.json ADDED
@@ -0,0 +1,1154 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 5.0,
6
+ "eval_steps": 500,
7
+ "global_step": 160,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.032,
14
+ "grad_norm": Infinity,
15
+ "learning_rate": 0.0,
16
+ "loss": 7.6104,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.064,
21
+ "grad_norm": Infinity,
22
+ "learning_rate": 0.0,
23
+ "loss": 7.5656,
24
+ "step": 2
25
+ },
26
+ {
27
+ "epoch": 0.096,
28
+ "grad_norm": 157.2535400390625,
29
+ "learning_rate": 0.0,
30
+ "loss": 7.6971,
31
+ "step": 3
32
+ },
33
+ {
34
+ "epoch": 0.128,
35
+ "grad_norm": 154.54010009765625,
36
+ "learning_rate": 1.25e-06,
37
+ "loss": 7.6309,
38
+ "step": 4
39
+ },
40
+ {
41
+ "epoch": 0.16,
42
+ "grad_norm": 152.9924774169922,
43
+ "learning_rate": 2.5e-06,
44
+ "loss": 7.3146,
45
+ "step": 5
46
+ },
47
+ {
48
+ "epoch": 0.192,
49
+ "grad_norm": 135.853271484375,
50
+ "learning_rate": 3.7500000000000005e-06,
51
+ "loss": 6.7531,
52
+ "step": 6
53
+ },
54
+ {
55
+ "epoch": 0.224,
56
+ "grad_norm": 98.87725830078125,
57
+ "learning_rate": 5e-06,
58
+ "loss": 5.3812,
59
+ "step": 7
60
+ },
61
+ {
62
+ "epoch": 0.256,
63
+ "grad_norm": 73.27239990234375,
64
+ "learning_rate": 6.25e-06,
65
+ "loss": 4.2198,
66
+ "step": 8
67
+ },
68
+ {
69
+ "epoch": 0.288,
70
+ "grad_norm": 49.881317138671875,
71
+ "learning_rate": 7.500000000000001e-06,
72
+ "loss": 3.3294,
73
+ "step": 9
74
+ },
75
+ {
76
+ "epoch": 0.32,
77
+ "grad_norm": 40.76887130737305,
78
+ "learning_rate": 8.750000000000001e-06,
79
+ "loss": 2.6914,
80
+ "step": 10
81
+ },
82
+ {
83
+ "epoch": 0.352,
84
+ "grad_norm": 37.55649185180664,
85
+ "learning_rate": 1e-05,
86
+ "loss": 2.1702,
87
+ "step": 11
88
+ },
89
+ {
90
+ "epoch": 0.384,
91
+ "grad_norm": 56.88014221191406,
92
+ "learning_rate": 9.998932083939657e-06,
93
+ "loss": 1.8042,
94
+ "step": 12
95
+ },
96
+ {
97
+ "epoch": 0.416,
98
+ "grad_norm": 28.132137298583984,
99
+ "learning_rate": 9.995728791936505e-06,
100
+ "loss": 1.6454,
101
+ "step": 13
102
+ },
103
+ {
104
+ "epoch": 0.448,
105
+ "grad_norm": 34.27433395385742,
106
+ "learning_rate": 9.990391492329341e-06,
107
+ "loss": 1.6005,
108
+ "step": 14
109
+ },
110
+ {
111
+ "epoch": 0.48,
112
+ "grad_norm": 56.15974807739258,
113
+ "learning_rate": 9.98292246503335e-06,
114
+ "loss": 1.6138,
115
+ "step": 15
116
+ },
117
+ {
118
+ "epoch": 0.512,
119
+ "grad_norm": 25.375675201416016,
120
+ "learning_rate": 9.973324900566214e-06,
121
+ "loss": 1.4571,
122
+ "step": 16
123
+ },
124
+ {
125
+ "epoch": 0.544,
126
+ "grad_norm": 26.252838134765625,
127
+ "learning_rate": 9.961602898685225e-06,
128
+ "loss": 1.4442,
129
+ "step": 17
130
+ },
131
+ {
132
+ "epoch": 0.576,
133
+ "grad_norm": 26.513261795043945,
134
+ "learning_rate": 9.947761466636014e-06,
135
+ "loss": 1.3839,
136
+ "step": 18
137
+ },
138
+ {
139
+ "epoch": 0.608,
140
+ "grad_norm": 26.487327575683594,
141
+ "learning_rate": 9.931806517013612e-06,
142
+ "loss": 1.3754,
143
+ "step": 19
144
+ },
145
+ {
146
+ "epoch": 0.64,
147
+ "grad_norm": 31.622228622436523,
148
+ "learning_rate": 9.913744865236798e-06,
149
+ "loss": 1.3904,
150
+ "step": 20
151
+ },
152
+ {
153
+ "epoch": 0.672,
154
+ "grad_norm": 27.828765869140625,
155
+ "learning_rate": 9.893584226636773e-06,
156
+ "loss": 1.3485,
157
+ "step": 21
158
+ },
159
+ {
160
+ "epoch": 0.704,
161
+ "grad_norm": 35.08741760253906,
162
+ "learning_rate": 9.871333213161438e-06,
163
+ "loss": 1.3843,
164
+ "step": 22
165
+ },
166
+ {
167
+ "epoch": 0.736,
168
+ "grad_norm": 29.33647346496582,
169
+ "learning_rate": 9.847001329696653e-06,
170
+ "loss": 1.2955,
171
+ "step": 23
172
+ },
173
+ {
174
+ "epoch": 0.768,
175
+ "grad_norm": 29.842397689819336,
176
+ "learning_rate": 9.820598970006068e-06,
177
+ "loss": 1.2918,
178
+ "step": 24
179
+ },
180
+ {
181
+ "epoch": 0.8,
182
+ "grad_norm": 31.41671371459961,
183
+ "learning_rate": 9.792137412291265e-06,
184
+ "loss": 1.2758,
185
+ "step": 25
186
+ },
187
+ {
188
+ "epoch": 0.832,
189
+ "grad_norm": 30.875192642211914,
190
+ "learning_rate": 9.761628814374074e-06,
191
+ "loss": 1.243,
192
+ "step": 26
193
+ },
194
+ {
195
+ "epoch": 0.864,
196
+ "grad_norm": 30.959537506103516,
197
+ "learning_rate": 9.729086208503174e-06,
198
+ "loss": 1.2242,
199
+ "step": 27
200
+ },
201
+ {
202
+ "epoch": 0.896,
203
+ "grad_norm": 31.773395538330078,
204
+ "learning_rate": 9.694523495787149e-06,
205
+ "loss": 1.2225,
206
+ "step": 28
207
+ },
208
+ {
209
+ "epoch": 0.928,
210
+ "grad_norm": 31.180255889892578,
211
+ "learning_rate": 9.657955440256396e-06,
212
+ "loss": 1.191,
213
+ "step": 29
214
+ },
215
+ {
216
+ "epoch": 0.96,
217
+ "grad_norm": 31.378707885742188,
218
+ "learning_rate": 9.619397662556434e-06,
219
+ "loss": 1.1779,
220
+ "step": 30
221
+ },
222
+ {
223
+ "epoch": 0.992,
224
+ "grad_norm": 31.57331085205078,
225
+ "learning_rate": 9.578866633275289e-06,
226
+ "loss": 1.1629,
227
+ "step": 31
228
+ },
229
+ {
230
+ "epoch": 1.0,
231
+ "grad_norm": Infinity,
232
+ "learning_rate": 9.536379665907801e-06,
233
+ "loss": 1.1432,
234
+ "step": 32
235
+ },
236
+ {
237
+ "epoch": 1.032,
238
+ "grad_norm": 31.775487899780273,
239
+ "learning_rate": 9.536379665907801e-06,
240
+ "loss": 1.1484,
241
+ "step": 33
242
+ },
243
+ {
244
+ "epoch": 1.064,
245
+ "grad_norm": 32.05231475830078,
246
+ "learning_rate": 9.491954909459895e-06,
247
+ "loss": 1.1361,
248
+ "step": 34
249
+ },
250
+ {
251
+ "epoch": 1.096,
252
+ "grad_norm": 32.09402084350586,
253
+ "learning_rate": 9.445611340695926e-06,
254
+ "loss": 1.1245,
255
+ "step": 35
256
+ },
257
+ {
258
+ "epoch": 1.1280000000000001,
259
+ "grad_norm": 32.11634063720703,
260
+ "learning_rate": 9.397368756032445e-06,
261
+ "loss": 1.1088,
262
+ "step": 36
263
+ },
264
+ {
265
+ "epoch": 1.16,
266
+ "grad_norm": 32.145057678222656,
267
+ "learning_rate": 9.347247763081834e-06,
268
+ "loss": 1.0929,
269
+ "step": 37
270
+ },
271
+ {
272
+ "epoch": 1.192,
273
+ "grad_norm": 32.16166687011719,
274
+ "learning_rate": 9.295269771849426e-06,
275
+ "loss": 1.0782,
276
+ "step": 38
277
+ },
278
+ {
279
+ "epoch": 1.224,
280
+ "grad_norm": 32.21006393432617,
281
+ "learning_rate": 9.241456985587868e-06,
282
+ "loss": 1.0666,
283
+ "step": 39
284
+ },
285
+ {
286
+ "epoch": 1.256,
287
+ "grad_norm": 32.27949523925781,
288
+ "learning_rate": 9.185832391312644e-06,
289
+ "loss": 1.0545,
290
+ "step": 40
291
+ },
292
+ {
293
+ "epoch": 1.288,
294
+ "grad_norm": 32.40619659423828,
295
+ "learning_rate": 9.12841974998278e-06,
296
+ "loss": 1.0413,
297
+ "step": 41
298
+ },
299
+ {
300
+ "epoch": 1.32,
301
+ "grad_norm": 32.46085739135742,
302
+ "learning_rate": 9.069243586350976e-06,
303
+ "loss": 1.0285,
304
+ "step": 42
305
+ },
306
+ {
307
+ "epoch": 1.3519999999999999,
308
+ "grad_norm": 32.50187683105469,
309
+ "learning_rate": 9.008329178487442e-06,
310
+ "loss": 1.0146,
311
+ "step": 43
312
+ },
313
+ {
314
+ "epoch": 1.384,
315
+ "grad_norm": 32.46120071411133,
316
+ "learning_rate": 8.94570254698197e-06,
317
+ "loss": 1.0024,
318
+ "step": 44
319
+ },
320
+ {
321
+ "epoch": 1.416,
322
+ "grad_norm": 32.52665328979492,
323
+ "learning_rate": 8.881390443828788e-06,
324
+ "loss": 0.9896,
325
+ "step": 45
326
+ },
327
+ {
328
+ "epoch": 1.448,
329
+ "grad_norm": 32.65945053100586,
330
+ "learning_rate": 8.815420340999034e-06,
331
+ "loss": 0.9801,
332
+ "step": 46
333
+ },
334
+ {
335
+ "epoch": 1.48,
336
+ "grad_norm": 32.593017578125,
337
+ "learning_rate": 8.747820418705632e-06,
338
+ "loss": 0.9662,
339
+ "step": 47
340
+ },
341
+ {
342
+ "epoch": 1.512,
343
+ "grad_norm": 32.5623779296875,
344
+ "learning_rate": 8.67861955336566e-06,
345
+ "loss": 0.9522,
346
+ "step": 48
347
+ },
348
+ {
349
+ "epoch": 1.544,
350
+ "grad_norm": 32.60811233520508,
351
+ "learning_rate": 8.607847305265312e-06,
352
+ "loss": 0.9393,
353
+ "step": 49
354
+ },
355
+ {
356
+ "epoch": 1.576,
357
+ "grad_norm": 32.59648895263672,
358
+ "learning_rate": 8.535533905932739e-06,
359
+ "loss": 0.9273,
360
+ "step": 50
361
+ },
362
+ {
363
+ "epoch": 1.608,
364
+ "grad_norm": 32.59210205078125,
365
+ "learning_rate": 8.461710245224149e-06,
366
+ "loss": 0.9161,
367
+ "step": 51
368
+ },
369
+ {
370
+ "epoch": 1.6400000000000001,
371
+ "grad_norm": 32.620819091796875,
372
+ "learning_rate": 8.386407858128707e-06,
373
+ "loss": 0.904,
374
+ "step": 52
375
+ },
376
+ {
377
+ "epoch": 1.6720000000000002,
378
+ "grad_norm": 32.551029205322266,
379
+ "learning_rate": 8.309658911297833e-06,
380
+ "loss": 0.8927,
381
+ "step": 53
382
+ },
383
+ {
384
+ "epoch": 1.704,
385
+ "grad_norm": 32.4809455871582,
386
+ "learning_rate": 8.231496189304704e-06,
387
+ "loss": 0.8783,
388
+ "step": 54
389
+ },
390
+ {
391
+ "epoch": 1.736,
392
+ "grad_norm": 32.426021575927734,
393
+ "learning_rate": 8.151953080639777e-06,
394
+ "loss": 0.8668,
395
+ "step": 55
396
+ },
397
+ {
398
+ "epoch": 1.768,
399
+ "grad_norm": 32.363037109375,
400
+ "learning_rate": 8.071063563448341e-06,
401
+ "loss": 0.8558,
402
+ "step": 56
403
+ },
404
+ {
405
+ "epoch": 1.8,
406
+ "grad_norm": 32.30210494995117,
407
+ "learning_rate": 7.988862191016204e-06,
408
+ "loss": 0.8434,
409
+ "step": 57
410
+ },
411
+ {
412
+ "epoch": 1.8319999999999999,
413
+ "grad_norm": 32.332271575927734,
414
+ "learning_rate": 7.905384077009693e-06,
415
+ "loss": 0.833,
416
+ "step": 58
417
+ },
418
+ {
419
+ "epoch": 1.8639999999999999,
420
+ "grad_norm": 32.23858642578125,
421
+ "learning_rate": 7.820664880476257e-06,
422
+ "loss": 0.8199,
423
+ "step": 59
424
+ },
425
+ {
426
+ "epoch": 1.896,
427
+ "grad_norm": 32.191062927246094,
428
+ "learning_rate": 7.734740790612137e-06,
429
+ "loss": 0.8077,
430
+ "step": 60
431
+ },
432
+ {
433
+ "epoch": 1.928,
434
+ "grad_norm": 32.13237762451172,
435
+ "learning_rate": 7.647648511303545e-06,
436
+ "loss": 0.797,
437
+ "step": 61
438
+ },
439
+ {
440
+ "epoch": 1.96,
441
+ "grad_norm": 32.073238372802734,
442
+ "learning_rate": 7.559425245448006e-06,
443
+ "loss": 0.7867,
444
+ "step": 62
445
+ },
446
+ {
447
+ "epoch": 1.992,
448
+ "grad_norm": 32.048255920410156,
449
+ "learning_rate": 7.470108679062521e-06,
450
+ "loss": 0.776,
451
+ "step": 63
452
+ },
453
+ {
454
+ "epoch": 2.0,
455
+ "grad_norm": Infinity,
456
+ "learning_rate": 7.379736965185369e-06,
457
+ "loss": 0.7636,
458
+ "step": 64
459
+ },
460
+ {
461
+ "epoch": 2.032,
462
+ "grad_norm": 32.028160095214844,
463
+ "learning_rate": 7.379736965185369e-06,
464
+ "loss": 0.7638,
465
+ "step": 65
466
+ },
467
+ {
468
+ "epoch": 2.064,
469
+ "grad_norm": 32.00226593017578,
470
+ "learning_rate": 7.288348707578409e-06,
471
+ "loss": 0.7528,
472
+ "step": 66
473
+ },
474
+ {
475
+ "epoch": 2.096,
476
+ "grad_norm": 31.98818016052246,
477
+ "learning_rate": 7.195982944236853e-06,
478
+ "loss": 0.7421,
479
+ "step": 67
480
+ },
481
+ {
482
+ "epoch": 2.128,
483
+ "grad_norm": 31.994173049926758,
484
+ "learning_rate": 7.102679130713538e-06,
485
+ "loss": 0.7303,
486
+ "step": 68
487
+ },
488
+ {
489
+ "epoch": 2.16,
490
+ "grad_norm": 31.960779190063477,
491
+ "learning_rate": 7.008477123264849e-06,
492
+ "loss": 0.7203,
493
+ "step": 69
494
+ },
495
+ {
496
+ "epoch": 2.192,
497
+ "grad_norm": 31.953826904296875,
498
+ "learning_rate": 6.913417161825449e-06,
499
+ "loss": 0.7105,
500
+ "step": 70
501
+ },
502
+ {
503
+ "epoch": 2.224,
504
+ "grad_norm": 31.943458557128906,
505
+ "learning_rate": 6.817539852819149e-06,
506
+ "loss": 0.7003,
507
+ "step": 71
508
+ },
509
+ {
510
+ "epoch": 2.2560000000000002,
511
+ "grad_norm": 31.915714263916016,
512
+ "learning_rate": 6.720886151813194e-06,
513
+ "loss": 0.6899,
514
+ "step": 72
515
+ },
516
+ {
517
+ "epoch": 2.288,
518
+ "grad_norm": 31.909622192382812,
519
+ "learning_rate": 6.6234973460234184e-06,
520
+ "loss": 0.6782,
521
+ "step": 73
522
+ },
523
+ {
524
+ "epoch": 2.32,
525
+ "grad_norm": 31.903554916381836,
526
+ "learning_rate": 6.525415036677745e-06,
527
+ "loss": 0.6689,
528
+ "step": 74
529
+ },
530
+ {
531
+ "epoch": 2.352,
532
+ "grad_norm": 31.868751525878906,
533
+ "learning_rate": 6.426681121245527e-06,
534
+ "loss": 0.6582,
535
+ "step": 75
536
+ },
537
+ {
538
+ "epoch": 2.384,
539
+ "grad_norm": 31.84931182861328,
540
+ "learning_rate": 6.327337775540362e-06,
541
+ "loss": 0.649,
542
+ "step": 76
543
+ },
544
+ {
545
+ "epoch": 2.416,
546
+ "grad_norm": 31.811887741088867,
547
+ "learning_rate": 6.227427435703997e-06,
548
+ "loss": 0.6405,
549
+ "step": 77
550
+ },
551
+ {
552
+ "epoch": 2.448,
553
+ "grad_norm": 31.750198364257812,
554
+ "learning_rate": 6.126992780079032e-06,
555
+ "loss": 0.6309,
556
+ "step": 78
557
+ },
558
+ {
559
+ "epoch": 2.48,
560
+ "grad_norm": 31.75521469116211,
561
+ "learning_rate": 6.026076710978172e-06,
562
+ "loss": 0.6222,
563
+ "step": 79
564
+ },
565
+ {
566
+ "epoch": 2.512,
567
+ "grad_norm": 31.68766212463379,
568
+ "learning_rate": 5.924722336357793e-06,
569
+ "loss": 0.6127,
570
+ "step": 80
571
+ },
572
+ {
573
+ "epoch": 2.544,
574
+ "grad_norm": 31.62371063232422,
575
+ "learning_rate": 5.82297295140367e-06,
576
+ "loss": 0.605,
577
+ "step": 81
578
+ },
579
+ {
580
+ "epoch": 2.576,
581
+ "grad_norm": 31.596736907958984,
582
+ "learning_rate": 5.720872020036734e-06,
583
+ "loss": 0.595,
584
+ "step": 82
585
+ },
586
+ {
587
+ "epoch": 2.608,
588
+ "grad_norm": 31.496021270751953,
589
+ "learning_rate": 5.61846315634674e-06,
590
+ "loss": 0.5872,
591
+ "step": 83
592
+ },
593
+ {
594
+ "epoch": 2.64,
595
+ "grad_norm": 31.44156265258789,
596
+ "learning_rate": 5.515790105961785e-06,
597
+ "loss": 0.5771,
598
+ "step": 84
599
+ },
600
+ {
601
+ "epoch": 2.672,
602
+ "grad_norm": 31.424474716186523,
603
+ "learning_rate": 5.412896727361663e-06,
604
+ "loss": 0.5686,
605
+ "step": 85
606
+ },
607
+ {
608
+ "epoch": 2.7039999999999997,
609
+ "grad_norm": 31.398405075073242,
610
+ "learning_rate": 5.309826973142974e-06,
611
+ "loss": 0.5615,
612
+ "step": 86
613
+ },
614
+ {
615
+ "epoch": 2.7359999999999998,
616
+ "grad_norm": 31.405683517456055,
617
+ "learning_rate": 5.206624871244066e-06,
618
+ "loss": 0.5545,
619
+ "step": 87
620
+ },
621
+ {
622
+ "epoch": 2.768,
623
+ "grad_norm": 31.285781860351562,
624
+ "learning_rate": 5.103334506137773e-06,
625
+ "loss": 0.5469,
626
+ "step": 88
627
+ },
628
+ {
629
+ "epoch": 2.8,
630
+ "grad_norm": 31.2457332611084,
631
+ "learning_rate": 5e-06,
632
+ "loss": 0.5411,
633
+ "step": 89
634
+ },
635
+ {
636
+ "epoch": 2.832,
637
+ "grad_norm": 31.182796478271484,
638
+ "learning_rate": 4.89666549386223e-06,
639
+ "loss": 0.5322,
640
+ "step": 90
641
+ },
642
+ {
643
+ "epoch": 2.864,
644
+ "grad_norm": 31.120420455932617,
645
+ "learning_rate": 4.793375128755934e-06,
646
+ "loss": 0.5264,
647
+ "step": 91
648
+ },
649
+ {
650
+ "epoch": 2.896,
651
+ "grad_norm": 31.106569290161133,
652
+ "learning_rate": 4.690173026857028e-06,
653
+ "loss": 0.5179,
654
+ "step": 92
655
+ },
656
+ {
657
+ "epoch": 2.928,
658
+ "grad_norm": 31.05204963684082,
659
+ "learning_rate": 4.587103272638339e-06,
660
+ "loss": 0.512,
661
+ "step": 93
662
+ },
663
+ {
664
+ "epoch": 2.96,
665
+ "grad_norm": 30.98640251159668,
666
+ "learning_rate": 4.4842098940382155e-06,
667
+ "loss": 0.5042,
668
+ "step": 94
669
+ },
670
+ {
671
+ "epoch": 2.992,
672
+ "grad_norm": 30.895837783813477,
673
+ "learning_rate": 4.381536843653262e-06,
674
+ "loss": 0.4975,
675
+ "step": 95
676
+ },
677
+ {
678
+ "epoch": 3.0,
679
+ "grad_norm": Infinity,
680
+ "learning_rate": 4.279127979963266e-06,
681
+ "loss": 0.4904,
682
+ "step": 96
683
+ },
684
+ {
685
+ "epoch": 3.032,
686
+ "grad_norm": 30.81951332092285,
687
+ "learning_rate": 4.279127979963266e-06,
688
+ "loss": 0.4912,
689
+ "step": 97
690
+ },
691
+ {
692
+ "epoch": 3.064,
693
+ "grad_norm": 30.759660720825195,
694
+ "learning_rate": 4.17702704859633e-06,
695
+ "loss": 0.4854,
696
+ "step": 98
697
+ },
698
+ {
699
+ "epoch": 3.096,
700
+ "grad_norm": 30.72711181640625,
701
+ "learning_rate": 4.075277663642208e-06,
702
+ "loss": 0.4799,
703
+ "step": 99
704
+ },
705
+ {
706
+ "epoch": 3.128,
707
+ "grad_norm": 30.67826271057129,
708
+ "learning_rate": 3.973923289021829e-06,
709
+ "loss": 0.4754,
710
+ "step": 100
711
+ },
712
+ {
713
+ "epoch": 3.16,
714
+ "grad_norm": 30.5875186920166,
715
+ "learning_rate": 3.8730072199209705e-06,
716
+ "loss": 0.4676,
717
+ "step": 101
718
+ },
719
+ {
720
+ "epoch": 3.192,
721
+ "grad_norm": 30.565122604370117,
722
+ "learning_rate": 3.7725725642960047e-06,
723
+ "loss": 0.4627,
724
+ "step": 102
725
+ },
726
+ {
727
+ "epoch": 3.224,
728
+ "grad_norm": 30.46950340270996,
729
+ "learning_rate": 3.67266222445964e-06,
730
+ "loss": 0.4574,
731
+ "step": 103
732
+ },
733
+ {
734
+ "epoch": 3.2560000000000002,
735
+ "grad_norm": 30.430648803710938,
736
+ "learning_rate": 3.573318878754475e-06,
737
+ "loss": 0.4524,
738
+ "step": 104
739
+ },
740
+ {
741
+ "epoch": 3.288,
742
+ "grad_norm": 30.337705612182617,
743
+ "learning_rate": 3.4745849633222566e-06,
744
+ "loss": 0.4476,
745
+ "step": 105
746
+ },
747
+ {
748
+ "epoch": 3.32,
749
+ "grad_norm": 30.253421783447266,
750
+ "learning_rate": 3.3765026539765832e-06,
751
+ "loss": 0.4434,
752
+ "step": 106
753
+ },
754
+ {
755
+ "epoch": 3.352,
756
+ "grad_norm": 30.180065155029297,
757
+ "learning_rate": 3.2791138481868084e-06,
758
+ "loss": 0.4389,
759
+ "step": 107
760
+ },
761
+ {
762
+ "epoch": 3.384,
763
+ "grad_norm": 30.156455993652344,
764
+ "learning_rate": 3.1824601471808504e-06,
765
+ "loss": 0.4327,
766
+ "step": 108
767
+ },
768
+ {
769
+ "epoch": 3.416,
770
+ "grad_norm": 30.073293685913086,
771
+ "learning_rate": 3.0865828381745515e-06,
772
+ "loss": 0.4282,
773
+ "step": 109
774
+ },
775
+ {
776
+ "epoch": 3.448,
777
+ "grad_norm": 30.003313064575195,
778
+ "learning_rate": 2.991522876735154e-06,
779
+ "loss": 0.4247,
780
+ "step": 110
781
+ },
782
+ {
783
+ "epoch": 3.48,
784
+ "grad_norm": 29.922578811645508,
785
+ "learning_rate": 2.8973208692864623e-06,
786
+ "loss": 0.4198,
787
+ "step": 111
788
+ },
789
+ {
790
+ "epoch": 3.512,
791
+ "grad_norm": 29.898513793945312,
792
+ "learning_rate": 2.804017055763149e-06,
793
+ "loss": 0.4165,
794
+ "step": 112
795
+ },
796
+ {
797
+ "epoch": 3.544,
798
+ "grad_norm": 29.841838836669922,
799
+ "learning_rate": 2.711651292421593e-06,
800
+ "loss": 0.4126,
801
+ "step": 113
802
+ },
803
+ {
804
+ "epoch": 3.576,
805
+ "grad_norm": 29.773988723754883,
806
+ "learning_rate": 2.6202630348146323e-06,
807
+ "loss": 0.409,
808
+ "step": 114
809
+ },
810
+ {
811
+ "epoch": 3.608,
812
+ "grad_norm": 29.704248428344727,
813
+ "learning_rate": 2.529891320937481e-06,
814
+ "loss": 0.4057,
815
+ "step": 115
816
+ },
817
+ {
818
+ "epoch": 3.64,
819
+ "grad_norm": 29.72831916809082,
820
+ "learning_rate": 2.4405747545519966e-06,
821
+ "loss": 0.4028,
822
+ "step": 116
823
+ },
824
+ {
825
+ "epoch": 3.672,
826
+ "grad_norm": 29.609638214111328,
827
+ "learning_rate": 2.352351488696457e-06,
828
+ "loss": 0.399,
829
+ "step": 117
830
+ },
831
+ {
832
+ "epoch": 3.7039999999999997,
833
+ "grad_norm": 29.58837127685547,
834
+ "learning_rate": 2.265259209387867e-06,
835
+ "loss": 0.3956,
836
+ "step": 118
837
+ },
838
+ {
839
+ "epoch": 3.7359999999999998,
840
+ "grad_norm": 29.477474212646484,
841
+ "learning_rate": 2.179335119523745e-06,
842
+ "loss": 0.3927,
843
+ "step": 119
844
+ },
845
+ {
846
+ "epoch": 3.768,
847
+ "grad_norm": 29.41726303100586,
848
+ "learning_rate": 2.094615922990309e-06,
849
+ "loss": 0.3897,
850
+ "step": 120
851
+ },
852
+ {
853
+ "epoch": 3.8,
854
+ "grad_norm": 29.370594024658203,
855
+ "learning_rate": 2.0111378089837958e-06,
856
+ "loss": 0.3866,
857
+ "step": 121
858
+ },
859
+ {
860
+ "epoch": 3.832,
861
+ "grad_norm": 29.35842514038086,
862
+ "learning_rate": 1.928936436551661e-06,
863
+ "loss": 0.3847,
864
+ "step": 122
865
+ },
866
+ {
867
+ "epoch": 3.864,
868
+ "grad_norm": 29.299884796142578,
869
+ "learning_rate": 1.848046919360225e-06,
870
+ "loss": 0.3821,
871
+ "step": 123
872
+ },
873
+ {
874
+ "epoch": 3.896,
875
+ "grad_norm": 29.2962703704834,
876
+ "learning_rate": 1.7685038106952952e-06,
877
+ "loss": 0.3799,
878
+ "step": 124
879
+ },
880
+ {
881
+ "epoch": 3.928,
882
+ "grad_norm": 29.24640464782715,
883
+ "learning_rate": 1.6903410887021676e-06,
884
+ "loss": 0.3777,
885
+ "step": 125
886
+ },
887
+ {
888
+ "epoch": 3.96,
889
+ "grad_norm": 29.137550354003906,
890
+ "learning_rate": 1.6135921418712959e-06,
891
+ "loss": 0.3749,
892
+ "step": 126
893
+ },
894
+ {
895
+ "epoch": 3.992,
896
+ "grad_norm": 29.106548309326172,
897
+ "learning_rate": 1.5382897547758513e-06,
898
+ "loss": 0.3734,
899
+ "step": 127
900
+ },
901
+ {
902
+ "epoch": 4.0,
903
+ "grad_norm": 29.02272605895996,
904
+ "learning_rate": 1.4644660940672628e-06,
905
+ "loss": 0.3718,
906
+ "step": 128
907
+ },
908
+ {
909
+ "epoch": 4.032,
910
+ "grad_norm": 29.036943435668945,
911
+ "learning_rate": 1.3921526947346902e-06,
912
+ "loss": 0.3698,
913
+ "step": 129
914
+ },
915
+ {
916
+ "epoch": 4.064,
917
+ "grad_norm": 29.032577514648438,
918
+ "learning_rate": 1.321380446634342e-06,
919
+ "loss": 0.368,
920
+ "step": 130
921
+ },
922
+ {
923
+ "epoch": 4.096,
924
+ "grad_norm": 29.02582359313965,
925
+ "learning_rate": 1.2521795812943704e-06,
926
+ "loss": 0.3667,
927
+ "step": 131
928
+ },
929
+ {
930
+ "epoch": 4.128,
931
+ "grad_norm": 28.966033935546875,
932
+ "learning_rate": 1.1845796590009684e-06,
933
+ "loss": 0.3646,
934
+ "step": 132
935
+ },
936
+ {
937
+ "epoch": 4.16,
938
+ "grad_norm": 28.905071258544922,
939
+ "learning_rate": 1.118609556171213e-06,
940
+ "loss": 0.3623,
941
+ "step": 133
942
+ },
943
+ {
944
+ "epoch": 4.192,
945
+ "grad_norm": 28.8817081451416,
946
+ "learning_rate": 1.0542974530180327e-06,
947
+ "loss": 0.3609,
948
+ "step": 134
949
+ },
950
+ {
951
+ "epoch": 4.224,
952
+ "grad_norm": 28.79871940612793,
953
+ "learning_rate": 9.916708215125586e-07,
954
+ "loss": 0.3588,
955
+ "step": 135
956
+ },
957
+ {
958
+ "epoch": 4.256,
959
+ "grad_norm": 28.76542091369629,
960
+ "learning_rate": 9.307564136490255e-07,
961
+ "loss": 0.3577,
962
+ "step": 136
963
+ },
964
+ {
965
+ "epoch": 4.288,
966
+ "grad_norm": 28.721641540527344,
967
+ "learning_rate": 8.715802500172215e-07,
968
+ "loss": 0.3562,
969
+ "step": 137
970
+ },
971
+ {
972
+ "epoch": 4.32,
973
+ "grad_norm": 28.721982955932617,
974
+ "learning_rate": 8.141676086873574e-07,
975
+ "loss": 0.3558,
976
+ "step": 138
977
+ },
978
+ {
979
+ "epoch": 4.352,
980
+ "grad_norm": 28.66592788696289,
981
+ "learning_rate": 7.585430144121319e-07,
982
+ "loss": 0.3545,
983
+ "step": 139
984
+ },
985
+ {
986
+ "epoch": 4.384,
987
+ "grad_norm": 28.666488647460938,
988
+ "learning_rate": 7.047302281505735e-07,
989
+ "loss": 0.3539,
990
+ "step": 140
991
+ },
992
+ {
993
+ "epoch": 4.416,
994
+ "grad_norm": 28.634742736816406,
995
+ "learning_rate": 6.527522369181655e-07,
996
+ "loss": 0.3527,
997
+ "step": 141
998
+ },
999
+ {
1000
+ "epoch": 4.448,
1001
+ "grad_norm": 28.640512466430664,
1002
+ "learning_rate": 6.026312439675553e-07,
1003
+ "loss": 0.3526,
1004
+ "step": 142
1005
+ },
1006
+ {
1007
+ "epoch": 4.48,
1008
+ "grad_norm": 28.617761611938477,
1009
+ "learning_rate": 5.543886593040737e-07,
1010
+ "loss": 0.3512,
1011
+ "step": 143
1012
+ },
1013
+ {
1014
+ "epoch": 4.5120000000000005,
1015
+ "grad_norm": 28.58139419555664,
1016
+ "learning_rate": 5.080450905401057e-07,
1017
+ "loss": 0.3504,
1018
+ "step": 144
1019
+ },
1020
+ {
1021
+ "epoch": 4.5440000000000005,
1022
+ "grad_norm": 28.599130630493164,
1023
+ "learning_rate": 4.6362033409220077e-07,
1024
+ "loss": 0.3507,
1025
+ "step": 145
1026
+ },
1027
+ {
1028
+ "epoch": 4.576,
1029
+ "grad_norm": 28.576433181762695,
1030
+ "learning_rate": 4.211333667247125e-07,
1031
+ "loss": 0.3496,
1032
+ "step": 146
1033
+ },
1034
+ {
1035
+ "epoch": 4.608,
1036
+ "grad_norm": 28.552490234375,
1037
+ "learning_rate": 3.8060233744356634e-07,
1038
+ "loss": 0.3478,
1039
+ "step": 147
1040
+ },
1041
+ {
1042
+ "epoch": 4.64,
1043
+ "grad_norm": 28.539840698242188,
1044
+ "learning_rate": 3.420445597436056e-07,
1045
+ "loss": 0.348,
1046
+ "step": 148
1047
+ },
1048
+ {
1049
+ "epoch": 4.672,
1050
+ "grad_norm": 28.509275436401367,
1051
+ "learning_rate": 3.0547650421285216e-07,
1052
+ "loss": 0.3466,
1053
+ "step": 149
1054
+ },
1055
+ {
1056
+ "epoch": 4.704,
1057
+ "grad_norm": 28.528343200683594,
1058
+ "learning_rate": 2.7091379149682683e-07,
1059
+ "loss": 0.3469,
1060
+ "step": 150
1061
+ },
1062
+ {
1063
+ "epoch": 4.736,
1064
+ "grad_norm": 28.52490234375,
1065
+ "learning_rate": 2.3837118562592799e-07,
1066
+ "loss": 0.347,
1067
+ "step": 151
1068
+ },
1069
+ {
1070
+ "epoch": 4.768,
1071
+ "grad_norm": 28.492401123046875,
1072
+ "learning_rate": 2.0786258770873647e-07,
1073
+ "loss": 0.3464,
1074
+ "step": 152
1075
+ },
1076
+ {
1077
+ "epoch": 4.8,
1078
+ "grad_norm": 28.494077682495117,
1079
+ "learning_rate": 1.7940102999393194e-07,
1080
+ "loss": 0.346,
1081
+ "step": 153
1082
+ },
1083
+ {
1084
+ "epoch": 4.832,
1085
+ "grad_norm": 28.514745712280273,
1086
+ "learning_rate": 1.5299867030334815e-07,
1087
+ "loss": 0.3465,
1088
+ "step": 154
1089
+ },
1090
+ {
1091
+ "epoch": 4.864,
1092
+ "grad_norm": 28.534265518188477,
1093
+ "learning_rate": 1.286667868385627e-07,
1094
+ "loss": 0.3461,
1095
+ "step": 155
1096
+ },
1097
+ {
1098
+ "epoch": 4.896,
1099
+ "grad_norm": 28.496475219726562,
1100
+ "learning_rate": 1.0641577336322761e-07,
1101
+ "loss": 0.3459,
1102
+ "step": 156
1103
+ },
1104
+ {
1105
+ "epoch": 4.928,
1106
+ "grad_norm": 28.47964859008789,
1107
+ "learning_rate": 8.625513476320291e-08,
1108
+ "loss": 0.346,
1109
+ "step": 157
1110
+ },
1111
+ {
1112
+ "epoch": 4.96,
1113
+ "grad_norm": 28.495473861694336,
1114
+ "learning_rate": 6.819348298638839e-08,
1115
+ "loss": 0.346,
1116
+ "step": 158
1117
+ },
1118
+ {
1119
+ "epoch": 4.992,
1120
+ "grad_norm": 28.4780330657959,
1121
+ "learning_rate": 5.223853336398632e-08,
1122
+ "loss": 0.3459,
1123
+ "step": 159
1124
+ },
1125
+ {
1126
+ "epoch": 5.0,
1127
+ "grad_norm": 28.506147384643555,
1128
+ "learning_rate": 3.839710131477492e-08,
1129
+ "loss": 0.3455,
1130
+ "step": 160
1131
+ }
1132
+ ],
1133
+ "logging_steps": 1,
1134
+ "max_steps": 160,
1135
+ "num_input_tokens_seen": 0,
1136
+ "num_train_epochs": 5,
1137
+ "save_steps": 500,
1138
+ "stateful_callbacks": {
1139
+ "TrainerControl": {
1140
+ "args": {
1141
+ "should_epoch_stop": false,
1142
+ "should_evaluate": false,
1143
+ "should_log": false,
1144
+ "should_save": true,
1145
+ "should_training_stop": true
1146
+ },
1147
+ "attributes": {}
1148
+ }
1149
+ },
1150
+ "total_flos": 5383533315883008.0,
1151
+ "train_batch_size": 6,
1152
+ "trial_name": null,
1153
+ "trial_params": null
1154
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:179870a0a59e477942539f2619fe597d755c774d04f76f8f7ab68ec8463a07ba
3
+ size 5496
vocab.json ADDED
The diff for this file is too large to render. See raw diff