Hugging Face's logo

NewstaR
/
orbita-tiny-0820

Safetensors
qwen3
Model card Files
xet
Community
marcuscedricridia commited on
Commit
ffd9a0c
·
verified ·
1 Parent(s): cffeacf

Upload folder using huggingface_hub

Browse files
Files changed (17) hide show
  1. .gitattributes +1 -0
  2. added_tokens.json +28 -0
  3. chat_template.jinja +8 -0
  4. config.json +62 -0
  5. generation_config.json +8 -0
  6. merges.txt +0 -0
  7. model.safetensors +3 -0
  8. optimizer.pt +3 -0
  9. rng_state.pth +3 -0
  10. scaler.pt +3 -0
  11. scheduler.pt +3 -0
  12. special_tokens_map.json +31 -0
  13. tokenizer.json +3 -0
  14. tokenizer_config.json +240 -0
  15. trainer_state.json +3772 -0
  16. training_args.bin +3 -0
  17. 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
+ }
chat_template.jinja ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {% for message in messages %}{% if message['role'] == 'user' %}{{'<|im_start|>user
2
+ ' + message['content'] + '<|im_end|>
3
+ '}}{% elif message['role'] == 'assistant' %}{{'<|im_start|>assistant
4
+ ' + message['content'] + '<|im_end|>
5
+ ' }}{% else %}{{ '<|im_start|>system
6
+ ' + message['content'] + '<|im_end|>
7
+ ' }}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant
8
+ ' }}{% endif %}
config.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "layer_types": [
14
+ "full_attention",
15
+ "full_attention",
16
+ "full_attention",
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention"
42
+ ],
43
+ "max_position_embeddings": 32768,
44
+ "max_window_layers": 28,
45
+ "model_type": "qwen3",
46
+ "num_attention_heads": 16,
47
+ "num_hidden_layers": 28,
48
+ "num_key_value_heads": 8,
49
+ "pad_token_id": 151654,
50
+ "rms_norm_eps": 1e-06,
51
+ "rope_scaling": null,
52
+ "rope_theta": 1000000,
53
+ "sliding_window": null,
54
+ "tie_word_embeddings": true,
55
+ "torch_dtype": "float32",
56
+ "transformers_version": "4.52.4",
57
+ "unsloth_fixed": true,
58
+ "unsloth_version": "2025.6.4",
59
+ "use_cache": true,
60
+ "use_sliding_window": false,
61
+ "vocab_size": 151936
62
+ }
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 sha256:1072f38c87d3b352cff1199ddb05d23783b7c6453bcc5d277713c73aa325e4c4
3
+ size 2384234968
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ecf96a63c81e623b1370113f016204329aba5650576936edde50f6c01ede5b11
3
+ size 2140048776
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e4ae31f3bd6abd5e088309ad57fa2e995bc6dd61c02221bc158a3d63e6ad1f06
3
+ size 14244
scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a1fed29054fe63ddb6ce23a65c0cb32be244a9a0dfb20585e48aaa7081dba16e
3
+ size 988
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c72c31b906df3c6694e9029c6cfe1d380d29dea29025d81471e32f461a88bff
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 +1,3772 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 2.0,
6
+ "eval_steps": 500,
7
+ "global_step": 534,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0037488284910965324,
14
+ "grad_norm": 3.1443588733673096,
15
+ "learning_rate": 0.0,
16
+ "loss": 0.7334,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.007497656982193065,
21
+ "grad_norm": 2.8891870975494385,
22
+ "learning_rate": 3.7037037037037036e-07,
23
+ "loss": 0.8449,
24
+ "step": 2
25
+ },
26
+ {
27
+ "epoch": 0.011246485473289597,
28
+ "grad_norm": 3.069650888442993,
29
+ "learning_rate": 7.407407407407407e-07,
30
+ "loss": 0.8849,
31
+ "step": 3
32
+ },
33
+ {
34
+ "epoch": 0.01499531396438613,
35
+ "grad_norm": 3.2767655849456787,
36
+ "learning_rate": 1.111111111111111e-06,
37
+ "loss": 0.918,
38
+ "step": 4
39
+ },
40
+ {
41
+ "epoch": 0.01874414245548266,
42
+ "grad_norm": 3.732121467590332,
43
+ "learning_rate": 1.4814814814814815e-06,
44
+ "loss": 0.9034,
45
+ "step": 5
46
+ },
47
+ {
48
+ "epoch": 0.022492970946579195,
49
+ "grad_norm": 3.3119568824768066,
50
+ "learning_rate": 1.8518518518518519e-06,
51
+ "loss": 0.8088,
52
+ "step": 6
53
+ },
54
+ {
55
+ "epoch": 0.026241799437675725,
56
+ "grad_norm": 2.831653118133545,
57
+ "learning_rate": 2.222222222222222e-06,
58
+ "loss": 0.7799,
59
+ "step": 7
60
+ },
61
+ {
62
+ "epoch": 0.02999062792877226,
63
+ "grad_norm": 2.624796152114868,
64
+ "learning_rate": 2.5925925925925925e-06,
65
+ "loss": 0.7546,
66
+ "step": 8
67
+ },
68
+ {
69
+ "epoch": 0.033739456419868794,
70
+ "grad_norm": 2.72420597076416,
71
+ "learning_rate": 2.962962962962963e-06,
72
+ "loss": 0.9903,
73
+ "step": 9
74
+ },
75
+ {
76
+ "epoch": 0.03748828491096532,
77
+ "grad_norm": 2.9389147758483887,
78
+ "learning_rate": 3.3333333333333333e-06,
79
+ "loss": 0.8715,
80
+ "step": 10
81
+ },
82
+ {
83
+ "epoch": 0.041237113402061855,
84
+ "grad_norm": 3.242018461227417,
85
+ "learning_rate": 3.7037037037037037e-06,
86
+ "loss": 1.1089,
87
+ "step": 11
88
+ },
89
+ {
90
+ "epoch": 0.04498594189315839,
91
+ "grad_norm": 3.1691386699676514,
92
+ "learning_rate": 4.074074074074074e-06,
93
+ "loss": 0.9839,
94
+ "step": 12
95
+ },
96
+ {
97
+ "epoch": 0.04873477038425492,
98
+ "grad_norm": 3.211369037628174,
99
+ "learning_rate": 4.444444444444444e-06,
100
+ "loss": 0.6889,
101
+ "step": 13
102
+ },
103
+ {
104
+ "epoch": 0.05248359887535145,
105
+ "grad_norm": 2.753861665725708,
106
+ "learning_rate": 4.814814814814815e-06,
107
+ "loss": 0.9228,
108
+ "step": 14
109
+ },
110
+ {
111
+ "epoch": 0.056232427366447985,
112
+ "grad_norm": 2.5858447551727295,
113
+ "learning_rate": 5.185185185185185e-06,
114
+ "loss": 0.6551,
115
+ "step": 15
116
+ },
117
+ {
118
+ "epoch": 0.05998125585754452,
119
+ "grad_norm": 2.531930446624756,
120
+ "learning_rate": 5.555555555555557e-06,
121
+ "loss": 0.7092,
122
+ "step": 16
123
+ },
124
+ {
125
+ "epoch": 0.06373008434864105,
126
+ "grad_norm": 3.0797035694122314,
127
+ "learning_rate": 5.925925925925926e-06,
128
+ "loss": 0.6336,
129
+ "step": 17
130
+ },
131
+ {
132
+ "epoch": 0.06747891283973759,
133
+ "grad_norm": 2.6957359313964844,
134
+ "learning_rate": 6.296296296296297e-06,
135
+ "loss": 0.7413,
136
+ "step": 18
137
+ },
138
+ {
139
+ "epoch": 0.07122774133083412,
140
+ "grad_norm": 2.9657933712005615,
141
+ "learning_rate": 6.666666666666667e-06,
142
+ "loss": 0.9917,
143
+ "step": 19
144
+ },
145
+ {
146
+ "epoch": 0.07497656982193064,
147
+ "grad_norm": 2.8018271923065186,
148
+ "learning_rate": 7.0370370370370375e-06,
149
+ "loss": 0.7026,
150
+ "step": 20
151
+ },
152
+ {
153
+ "epoch": 0.07872539831302718,
154
+ "grad_norm": 3.1438984870910645,
155
+ "learning_rate": 7.4074074074074075e-06,
156
+ "loss": 1.0012,
157
+ "step": 21
158
+ },
159
+ {
160
+ "epoch": 0.08247422680412371,
161
+ "grad_norm": 2.5763065814971924,
162
+ "learning_rate": 7.77777777777778e-06,
163
+ "loss": 0.7979,
164
+ "step": 22
165
+ },
166
+ {
167
+ "epoch": 0.08622305529522024,
168
+ "grad_norm": 2.6274800300598145,
169
+ "learning_rate": 8.148148148148148e-06,
170
+ "loss": 0.6811,
171
+ "step": 23
172
+ },
173
+ {
174
+ "epoch": 0.08997188378631678,
175
+ "grad_norm": 2.6588563919067383,
176
+ "learning_rate": 8.518518518518519e-06,
177
+ "loss": 0.7003,
178
+ "step": 24
179
+ },
180
+ {
181
+ "epoch": 0.09372071227741331,
182
+ "grad_norm": 2.8751745223999023,
183
+ "learning_rate": 8.888888888888888e-06,
184
+ "loss": 1.1499,
185
+ "step": 25
186
+ },
187
+ {
188
+ "epoch": 0.09746954076850985,
189
+ "grad_norm": 2.9690346717834473,
190
+ "learning_rate": 9.25925925925926e-06,
191
+ "loss": 0.8019,
192
+ "step": 26
193
+ },
194
+ {
195
+ "epoch": 0.10121836925960637,
196
+ "grad_norm": 2.8583548069000244,
197
+ "learning_rate": 9.62962962962963e-06,
198
+ "loss": 0.926,
199
+ "step": 27
200
+ },
201
+ {
202
+ "epoch": 0.1049671977507029,
203
+ "grad_norm": 2.719120502471924,
204
+ "learning_rate": 1e-05,
205
+ "loss": 0.8262,
206
+ "step": 28
207
+ },
208
+ {
209
+ "epoch": 0.10871602624179943,
210
+ "grad_norm": 2.5264711380004883,
211
+ "learning_rate": 9.999904010783725e-06,
212
+ "loss": 0.7492,
213
+ "step": 29
214
+ },
215
+ {
216
+ "epoch": 0.11246485473289597,
217
+ "grad_norm": 2.5457921028137207,
218
+ "learning_rate": 9.999616046820467e-06,
219
+ "loss": 0.8453,
220
+ "step": 30
221
+ },
222
+ {
223
+ "epoch": 0.1162136832239925,
224
+ "grad_norm": 2.66349720954895,
225
+ "learning_rate": 9.999136119166803e-06,
226
+ "loss": 0.6645,
227
+ "step": 31
228
+ },
229
+ {
230
+ "epoch": 0.11996251171508904,
231
+ "grad_norm": 2.695261240005493,
232
+ "learning_rate": 9.998464246249885e-06,
233
+ "loss": 1.0498,
234
+ "step": 32
235
+ },
236
+ {
237
+ "epoch": 0.12371134020618557,
238
+ "grad_norm": 2.371567487716675,
239
+ "learning_rate": 9.997600453866734e-06,
240
+ "loss": 0.9992,
241
+ "step": 33
242
+ },
243
+ {
244
+ "epoch": 0.1274601686972821,
245
+ "grad_norm": 2.551347255706787,
246
+ "learning_rate": 9.99654477518325e-06,
247
+ "loss": 0.786,
248
+ "step": 34
249
+ },
250
+ {
251
+ "epoch": 0.13120899718837864,
252
+ "grad_norm": 2.8569281101226807,
253
+ "learning_rate": 9.995297250732942e-06,
254
+ "loss": 0.7571,
255
+ "step": 35
256
+ },
257
+ {
258
+ "epoch": 0.13495782567947517,
259
+ "grad_norm": 2.9101366996765137,
260
+ "learning_rate": 9.99385792841537e-06,
261
+ "loss": 0.8682,
262
+ "step": 36
263
+ },
264
+ {
265
+ "epoch": 0.1387066541705717,
266
+ "grad_norm": 2.8088221549987793,
267
+ "learning_rate": 9.9922268634943e-06,
268
+ "loss": 0.9882,
269
+ "step": 37
270
+ },
271
+ {
272
+ "epoch": 0.14245548266166824,
273
+ "grad_norm": 2.6386826038360596,
274
+ "learning_rate": 9.99040411859559e-06,
275
+ "loss": 0.8706,
276
+ "step": 38
277
+ },
278
+ {
279
+ "epoch": 0.14620431115276475,
280
+ "grad_norm": 2.306828022003174,
281
+ "learning_rate": 9.98838976370478e-06,
282
+ "loss": 0.7294,
283
+ "step": 39
284
+ },
285
+ {
286
+ "epoch": 0.14995313964386128,
287
+ "grad_norm": 3.0453381538391113,
288
+ "learning_rate": 9.986183876164412e-06,
289
+ "loss": 0.7128,
290
+ "step": 40
291
+ },
292
+ {
293
+ "epoch": 0.15370196813495782,
294
+ "grad_norm": 2.72247314453125,
295
+ "learning_rate": 9.983786540671052e-06,
296
+ "loss": 0.8392,
297
+ "step": 41
298
+ },
299
+ {
300
+ "epoch": 0.15745079662605435,
301
+ "grad_norm": 2.6883301734924316,
302
+ "learning_rate": 9.981197849272039e-06,
303
+ "loss": 0.6254,
304
+ "step": 42
305
+ },
306
+ {
307
+ "epoch": 0.16119962511715089,
308
+ "grad_norm": 2.640571355819702,
309
+ "learning_rate": 9.978417901361958e-06,
310
+ "loss": 0.7416,
311
+ "step": 43
312
+ },
313
+ {
314
+ "epoch": 0.16494845360824742,
315
+ "grad_norm": 2.496026039123535,
316
+ "learning_rate": 9.975446803678818e-06,
317
+ "loss": 0.9206,
318
+ "step": 44
319
+ },
320
+ {
321
+ "epoch": 0.16869728209934395,
322
+ "grad_norm": 2.7780568599700928,
323
+ "learning_rate": 9.972284670299957e-06,
324
+ "loss": 0.8848,
325
+ "step": 45
326
+ },
327
+ {
328
+ "epoch": 0.1724461105904405,
329
+ "grad_norm": 2.521110773086548,
330
+ "learning_rate": 9.968931622637652e-06,
331
+ "loss": 0.6183,
332
+ "step": 46
333
+ },
334
+ {
335
+ "epoch": 0.17619493908153702,
336
+ "grad_norm": 2.378988742828369,
337
+ "learning_rate": 9.965387789434474e-06,
338
+ "loss": 0.7197,
339
+ "step": 47
340
+ },
341
+ {
342
+ "epoch": 0.17994376757263356,
343
+ "grad_norm": 2.662712574005127,
344
+ "learning_rate": 9.961653306758326e-06,
345
+ "loss": 0.8191,
346
+ "step": 48
347
+ },
348
+ {
349
+ "epoch": 0.1836925960637301,
350
+ "grad_norm": 2.818058729171753,
351
+ "learning_rate": 9.95772831799724e-06,
352
+ "loss": 0.945,
353
+ "step": 49
354
+ },
355
+ {
356
+ "epoch": 0.18744142455482662,
357
+ "grad_norm": 2.4332478046417236,
358
+ "learning_rate": 9.953612973853853e-06,
359
+ "loss": 0.743,
360
+ "step": 50
361
+ },
362
+ {
363
+ "epoch": 0.19119025304592316,
364
+ "grad_norm": 2.722165584564209,
365
+ "learning_rate": 9.949307432339625e-06,
366
+ "loss": 0.7276,
367
+ "step": 51
368
+ },
369
+ {
370
+ "epoch": 0.1949390815370197,
371
+ "grad_norm": 3.4195361137390137,
372
+ "learning_rate": 9.944811858768782e-06,
373
+ "loss": 0.9687,
374
+ "step": 52
375
+ },
376
+ {
377
+ "epoch": 0.19868791002811623,
378
+ "grad_norm": 3.273954391479492,
379
+ "learning_rate": 9.940126425751957e-06,
380
+ "loss": 0.8701,
381
+ "step": 53
382
+ },
383
+ {
384
+ "epoch": 0.20243673851921273,
385
+ "grad_norm": 2.8895037174224854,
386
+ "learning_rate": 9.935251313189564e-06,
387
+ "loss": 1.1019,
388
+ "step": 54
389
+ },
390
+ {
391
+ "epoch": 0.20618556701030927,
392
+ "grad_norm": 2.51041316986084,
393
+ "learning_rate": 9.930186708264902e-06,
394
+ "loss": 0.6458,
395
+ "step": 55
396
+ },
397
+ {
398
+ "epoch": 0.2099343955014058,
399
+ "grad_norm": 2.807650566101074,
400
+ "learning_rate": 9.92493280543695e-06,
401
+ "loss": 0.7822,
402
+ "step": 56
403
+ },
404
+ {
405
+ "epoch": 0.21368322399250234,
406
+ "grad_norm": 2.7724640369415283,
407
+ "learning_rate": 9.919489806432915e-06,
408
+ "loss": 0.9093,
409
+ "step": 57
410
+ },
411
+ {
412
+ "epoch": 0.21743205248359887,
413
+ "grad_norm": 2.830974578857422,
414
+ "learning_rate": 9.913857920240481e-06,
415
+ "loss": 1.0478,
416
+ "step": 58
417
+ },
418
+ {
419
+ "epoch": 0.2211808809746954,
420
+ "grad_norm": 2.7909905910491943,
421
+ "learning_rate": 9.908037363099782e-06,
422
+ "loss": 0.6055,
423
+ "step": 59
424
+ },
425
+ {
426
+ "epoch": 0.22492970946579194,
427
+ "grad_norm": 2.5817978382110596,
428
+ "learning_rate": 9.90202835849511e-06,
429
+ "loss": 0.5351,
430
+ "step": 60
431
+ },
432
+ {
433
+ "epoch": 0.22867853795688847,
434
+ "grad_norm": 2.8909380435943604,
435
+ "learning_rate": 9.895831137146319e-06,
436
+ "loss": 0.7823,
437
+ "step": 61
438
+ },
439
+ {
440
+ "epoch": 0.232427366447985,
441
+ "grad_norm": 2.5689468383789062,
442
+ "learning_rate": 9.889445936999978e-06,
443
+ "loss": 0.7244,
444
+ "step": 62
445
+ },
446
+ {
447
+ "epoch": 0.23617619493908154,
448
+ "grad_norm": 2.3992927074432373,
449
+ "learning_rate": 9.882873003220229e-06,
450
+ "loss": 0.6762,
451
+ "step": 63
452
+ },
453
+ {
454
+ "epoch": 0.23992502343017807,
455
+ "grad_norm": 2.6709885597229004,
456
+ "learning_rate": 9.876112588179378e-06,
457
+ "loss": 1.0162,
458
+ "step": 64
459
+ },
460
+ {
461
+ "epoch": 0.2436738519212746,
462
+ "grad_norm": 2.898042917251587,
463
+ "learning_rate": 9.869164951448201e-06,
464
+ "loss": 0.8213,
465
+ "step": 65
466
+ },
467
+ {
468
+ "epoch": 0.24742268041237114,
469
+ "grad_norm": 2.4519455432891846,
470
+ "learning_rate": 9.86203035978598e-06,
471
+ "loss": 0.6992,
472
+ "step": 66
473
+ },
474
+ {
475
+ "epoch": 0.2511715089034677,
476
+ "grad_norm": 2.6127779483795166,
477
+ "learning_rate": 9.854709087130261e-06,
478
+ "loss": 0.7047,
479
+ "step": 67
480
+ },
481
+ {
482
+ "epoch": 0.2549203373945642,
483
+ "grad_norm": 2.591939926147461,
484
+ "learning_rate": 9.847201414586331e-06,
485
+ "loss": 0.8114,
486
+ "step": 68
487
+ },
488
+ {
489
+ "epoch": 0.25866916588566075,
490
+ "grad_norm": 2.322962522506714,
491
+ "learning_rate": 9.839507630416436e-06,
492
+ "loss": 0.567,
493
+ "step": 69
494
+ },
495
+ {
496
+ "epoch": 0.2624179943767573,
497
+ "grad_norm": 2.7526068687438965,
498
+ "learning_rate": 9.831628030028698e-06,
499
+ "loss": 0.8195,
500
+ "step": 70
501
+ },
502
+ {
503
+ "epoch": 0.2661668228678538,
504
+ "grad_norm": 3.0277445316314697,
505
+ "learning_rate": 9.82356291596578e-06,
506
+ "loss": 1.1821,
507
+ "step": 71
508
+ },
509
+ {
510
+ "epoch": 0.26991565135895035,
511
+ "grad_norm": 2.836676836013794,
512
+ "learning_rate": 9.81531259789328e-06,
513
+ "loss": 1.1377,
514
+ "step": 72
515
+ },
516
+ {
517
+ "epoch": 0.2736644798500469,
518
+ "grad_norm": 2.7574238777160645,
519
+ "learning_rate": 9.80687739258782e-06,
520
+ "loss": 0.8165,
521
+ "step": 73
522
+ },
523
+ {
524
+ "epoch": 0.2774133083411434,
525
+ "grad_norm": 2.917651414871216,
526
+ "learning_rate": 9.7982576239249e-06,
527
+ "loss": 0.8276,
528
+ "step": 74
529
+ },
530
+ {
531
+ "epoch": 0.28116213683223995,
532
+ "grad_norm": 2.5589005947113037,
533
+ "learning_rate": 9.789453622866455e-06,
534
+ "loss": 0.7906,
535
+ "step": 75
536
+ },
537
+ {
538
+ "epoch": 0.2849109653233365,
539
+ "grad_norm": 2.742344379425049,
540
+ "learning_rate": 9.78046572744815e-06,
541
+ "loss": 0.9195,
542
+ "step": 76
543
+ },
544
+ {
545
+ "epoch": 0.28865979381443296,
546
+ "grad_norm": 2.717301845550537,
547
+ "learning_rate": 9.771294282766399e-06,
548
+ "loss": 0.9222,
549
+ "step": 77
550
+ },
551
+ {
552
+ "epoch": 0.2924086223055295,
553
+ "grad_norm": 2.7094359397888184,
554
+ "learning_rate": 9.761939640965117e-06,
555
+ "loss": 0.8209,
556
+ "step": 78
557
+ },
558
+ {
559
+ "epoch": 0.29615745079662603,
560
+ "grad_norm": 2.3744776248931885,
561
+ "learning_rate": 9.7524021612222e-06,
562
+ "loss": 0.8834,
563
+ "step": 79
564
+ },
565
+ {
566
+ "epoch": 0.29990627928772257,
567
+ "grad_norm": 2.6806893348693848,
568
+ "learning_rate": 9.742682209735727e-06,
569
+ "loss": 0.6742,
570
+ "step": 80
571
+ },
572
+ {
573
+ "epoch": 0.3036551077788191,
574
+ "grad_norm": 2.5327813625335693,
575
+ "learning_rate": 9.732780159709912e-06,
576
+ "loss": 0.8571,
577
+ "step": 81
578
+ },
579
+ {
580
+ "epoch": 0.30740393626991563,
581
+ "grad_norm": 2.423889636993408,
582
+ "learning_rate": 9.722696391340762e-06,
583
+ "loss": 0.6847,
584
+ "step": 82
585
+ },
586
+ {
587
+ "epoch": 0.31115276476101217,
588
+ "grad_norm": 3.3473219871520996,
589
+ "learning_rate": 9.712431291801483e-06,
590
+ "loss": 0.997,
591
+ "step": 83
592
+ },
593
+ {
594
+ "epoch": 0.3149015932521087,
595
+ "grad_norm": 2.3630363941192627,
596
+ "learning_rate": 9.701985255227624e-06,
597
+ "loss": 0.7263,
598
+ "step": 84
599
+ },
600
+ {
601
+ "epoch": 0.31865042174320524,
602
+ "grad_norm": 2.9431068897247314,
603
+ "learning_rate": 9.691358682701927e-06,
604
+ "loss": 0.9122,
605
+ "step": 85
606
+ },
607
+ {
608
+ "epoch": 0.32239925023430177,
609
+ "grad_norm": 2.4724693298339844,
610
+ "learning_rate": 9.680551982238941e-06,
611
+ "loss": 0.85,
612
+ "step": 86
613
+ },
614
+ {
615
+ "epoch": 0.3261480787253983,
616
+ "grad_norm": 2.641633987426758,
617
+ "learning_rate": 9.669565568769348e-06,
618
+ "loss": 0.7426,
619
+ "step": 87
620
+ },
621
+ {
622
+ "epoch": 0.32989690721649484,
623
+ "grad_norm": 2.5549933910369873,
624
+ "learning_rate": 9.658399864124037e-06,
625
+ "loss": 0.7846,
626
+ "step": 88
627
+ },
628
+ {
629
+ "epoch": 0.3336457357075914,
630
+ "grad_norm": 2.469627618789673,
631
+ "learning_rate": 9.647055297017901e-06,
632
+ "loss": 0.745,
633
+ "step": 89
634
+ },
635
+ {
636
+ "epoch": 0.3373945641986879,
637
+ "grad_norm": 2.484022378921509,
638
+ "learning_rate": 9.635532303033386e-06,
639
+ "loss": 0.7989,
640
+ "step": 90
641
+ },
642
+ {
643
+ "epoch": 0.34114339268978444,
644
+ "grad_norm": 2.852081060409546,
645
+ "learning_rate": 9.623831324603755e-06,
646
+ "loss": 0.6607,
647
+ "step": 91
648
+ },
649
+ {
650
+ "epoch": 0.344892221180881,
651
+ "grad_norm": 2.902130126953125,
652
+ "learning_rate": 9.611952810996104e-06,
653
+ "loss": 0.8182,
654
+ "step": 92
655
+ },
656
+ {
657
+ "epoch": 0.3486410496719775,
658
+ "grad_norm": 2.2227494716644287,
659
+ "learning_rate": 9.599897218294122e-06,
660
+ "loss": 0.6317,
661
+ "step": 93
662
+ },
663
+ {
664
+ "epoch": 0.35238987816307404,
665
+ "grad_norm": 2.668668508529663,
666
+ "learning_rate": 9.587665009380565e-06,
667
+ "loss": 0.8303,
668
+ "step": 94
669
+ },
670
+ {
671
+ "epoch": 0.3561387066541706,
672
+ "grad_norm": 2.8259904384613037,
673
+ "learning_rate": 9.575256653919494e-06,
674
+ "loss": 0.8235,
675
+ "step": 95
676
+ },
677
+ {
678
+ "epoch": 0.3598875351452671,
679
+ "grad_norm": 2.3626370429992676,
680
+ "learning_rate": 9.562672628338233e-06,
681
+ "loss": 0.8566,
682
+ "step": 96
683
+ },
684
+ {
685
+ "epoch": 0.36363636363636365,
686
+ "grad_norm": 2.487170934677124,
687
+ "learning_rate": 9.549913415809084e-06,
688
+ "loss": 0.8119,
689
+ "step": 97
690
+ },
691
+ {
692
+ "epoch": 0.3673851921274602,
693
+ "grad_norm": 2.4403631687164307,
694
+ "learning_rate": 9.536979506230772e-06,
695
+ "loss": 0.6435,
696
+ "step": 98
697
+ },
698
+ {
699
+ "epoch": 0.3711340206185567,
700
+ "grad_norm": 2.5403356552124023,
701
+ "learning_rate": 9.523871396209633e-06,
702
+ "loss": 0.8494,
703
+ "step": 99
704
+ },
705
+ {
706
+ "epoch": 0.37488284910965325,
707
+ "grad_norm": 2.5626585483551025,
708
+ "learning_rate": 9.510589589040554e-06,
709
+ "loss": 0.8538,
710
+ "step": 100
711
+ },
712
+ {
713
+ "epoch": 0.3786316776007498,
714
+ "grad_norm": 2.8327720165252686,
715
+ "learning_rate": 9.497134594687635e-06,
716
+ "loss": 0.8465,
717
+ "step": 101
718
+ },
719
+ {
720
+ "epoch": 0.3823805060918463,
721
+ "grad_norm": 2.6077094078063965,
722
+ "learning_rate": 9.483506929764623e-06,
723
+ "loss": 0.9154,
724
+ "step": 102
725
+ },
726
+ {
727
+ "epoch": 0.38612933458294285,
728
+ "grad_norm": 2.6204569339752197,
729
+ "learning_rate": 9.469707117515068e-06,
730
+ "loss": 0.8607,
731
+ "step": 103
732
+ },
733
+ {
734
+ "epoch": 0.3898781630740394,
735
+ "grad_norm": 2.281536340713501,
736
+ "learning_rate": 9.455735687792233e-06,
737
+ "loss": 0.6667,
738
+ "step": 104
739
+ },
740
+ {
741
+ "epoch": 0.3936269915651359,
742
+ "grad_norm": 2.339271068572998,
743
+ "learning_rate": 9.44159317703876e-06,
744
+ "loss": 0.7999,
745
+ "step": 105
746
+ },
747
+ {
748
+ "epoch": 0.39737582005623245,
749
+ "grad_norm": 2.6674795150756836,
750
+ "learning_rate": 9.427280128266049e-06,
751
+ "loss": 0.8966,
752
+ "step": 106
753
+ },
754
+ {
755
+ "epoch": 0.40112464854732893,
756
+ "grad_norm": 2.6626410484313965,
757
+ "learning_rate": 9.412797091033444e-06,
758
+ "loss": 0.9888,
759
+ "step": 107
760
+ },
761
+ {
762
+ "epoch": 0.40487347703842547,
763
+ "grad_norm": 2.451646327972412,
764
+ "learning_rate": 9.398144621427095e-06,
765
+ "loss": 0.7585,
766
+ "step": 108
767
+ },
768
+ {
769
+ "epoch": 0.408622305529522,
770
+ "grad_norm": 2.4723331928253174,
771
+ "learning_rate": 9.383323282038632e-06,
772
+ "loss": 0.7715,
773
+ "step": 109
774
+ },
775
+ {
776
+ "epoch": 0.41237113402061853,
777
+ "grad_norm": 3.4575955867767334,
778
+ "learning_rate": 9.368333641943558e-06,
779
+ "loss": 0.6064,
780
+ "step": 110
781
+ },
782
+ {
783
+ "epoch": 0.41611996251171507,
784
+ "grad_norm": 2.347303867340088,
785
+ "learning_rate": 9.353176276679397e-06,
786
+ "loss": 0.8154,
787
+ "step": 111
788
+ },
789
+ {
790
+ "epoch": 0.4198687910028116,
791
+ "grad_norm": 2.850472927093506,
792
+ "learning_rate": 9.337851768223589e-06,
793
+ "loss": 0.8903,
794
+ "step": 112
795
+ },
796
+ {
797
+ "epoch": 0.42361761949390814,
798
+ "grad_norm": 2.8391191959381104,
799
+ "learning_rate": 9.322360704971161e-06,
800
+ "loss": 0.8927,
801
+ "step": 113
802
+ },
803
+ {
804
+ "epoch": 0.42736644798500467,
805
+ "grad_norm": 2.2651166915893555,
806
+ "learning_rate": 9.30670368171212e-06,
807
+ "loss": 0.5307,
808
+ "step": 114
809
+ },
810
+ {
811
+ "epoch": 0.4311152764761012,
812
+ "grad_norm": 2.3141026496887207,
813
+ "learning_rate": 9.29088129960862e-06,
814
+ "loss": 0.7067,
815
+ "step": 115
816
+ },
817
+ {
818
+ "epoch": 0.43486410496719774,
819
+ "grad_norm": 2.7184128761291504,
820
+ "learning_rate": 9.274894166171888e-06,
821
+ "loss": 1.0238,
822
+ "step": 116
823
+ },
824
+ {
825
+ "epoch": 0.4386129334582943,
826
+ "grad_norm": 2.4955923557281494,
827
+ "learning_rate": 9.258742895238886e-06,
828
+ "loss": 0.7917,
829
+ "step": 117
830
+ },
831
+ {
832
+ "epoch": 0.4423617619493908,
833
+ "grad_norm": 2.485121726989746,
834
+ "learning_rate": 9.242428106948748e-06,
835
+ "loss": 0.9189,
836
+ "step": 118
837
+ },
838
+ {
839
+ "epoch": 0.44611059044048734,
840
+ "grad_norm": 2.644423723220825,
841
+ "learning_rate": 9.225950427718974e-06,
842
+ "loss": 0.8239,
843
+ "step": 119
844
+ },
845
+ {
846
+ "epoch": 0.4498594189315839,
847
+ "grad_norm": 2.267098903656006,
848
+ "learning_rate": 9.209310490221368e-06,
849
+ "loss": 0.5885,
850
+ "step": 120
851
+ },
852
+ {
853
+ "epoch": 0.4536082474226804,
854
+ "grad_norm": 3.184373140335083,
855
+ "learning_rate": 9.192508933357753e-06,
856
+ "loss": 0.9993,
857
+ "step": 121
858
+ },
859
+ {
860
+ "epoch": 0.45735707591377694,
861
+ "grad_norm": 2.420222282409668,
862
+ "learning_rate": 9.175546402235443e-06,
863
+ "loss": 0.6963,
864
+ "step": 122
865
+ },
866
+ {
867
+ "epoch": 0.4611059044048735,
868
+ "grad_norm": 2.1986594200134277,
869
+ "learning_rate": 9.158423548142459e-06,
870
+ "loss": 0.6484,
871
+ "step": 123
872
+ },
873
+ {
874
+ "epoch": 0.46485473289597,
875
+ "grad_norm": 2.687732219696045,
876
+ "learning_rate": 9.141141028522544e-06,
877
+ "loss": 0.8614,
878
+ "step": 124
879
+ },
880
+ {
881
+ "epoch": 0.46860356138706655,
882
+ "grad_norm": 2.5353922843933105,
883
+ "learning_rate": 9.123699506949903e-06,
884
+ "loss": 0.901,
885
+ "step": 125
886
+ },
887
+ {
888
+ "epoch": 0.4723523898781631,
889
+ "grad_norm": 2.416156768798828,
890
+ "learning_rate": 9.106099653103729e-06,
891
+ "loss": 0.6771,
892
+ "step": 126
893
+ },
894
+ {
895
+ "epoch": 0.4761012183692596,
896
+ "grad_norm": 2.7599902153015137,
897
+ "learning_rate": 9.088342142742493e-06,
898
+ "loss": 0.6258,
899
+ "step": 127
900
+ },
901
+ {
902
+ "epoch": 0.47985004686035615,
903
+ "grad_norm": 2.6150290966033936,
904
+ "learning_rate": 9.070427657677996e-06,
905
+ "loss": 0.923,
906
+ "step": 128
907
+ },
908
+ {
909
+ "epoch": 0.4835988753514527,
910
+ "grad_norm": 2.746690034866333,
911
+ "learning_rate": 9.052356885749191e-06,
912
+ "loss": 0.7873,
913
+ "step": 129
914
+ },
915
+ {
916
+ "epoch": 0.4873477038425492,
917
+ "grad_norm": 2.5986545085906982,
918
+ "learning_rate": 9.034130520795774e-06,
919
+ "loss": 0.6669,
920
+ "step": 130
921
+ },
922
+ {
923
+ "epoch": 0.49109653233364575,
924
+ "grad_norm": 2.674226760864258,
925
+ "learning_rate": 9.015749262631537e-06,
926
+ "loss": 0.8936,
927
+ "step": 131
928
+ },
929
+ {
930
+ "epoch": 0.4948453608247423,
931
+ "grad_norm": 2.7541725635528564,
932
+ "learning_rate": 8.997213817017508e-06,
933
+ "loss": 0.9551,
934
+ "step": 132
935
+ },
936
+ {
937
+ "epoch": 0.4985941893158388,
938
+ "grad_norm": 2.3283989429473877,
939
+ "learning_rate": 8.978524895634842e-06,
940
+ "loss": 0.8066,
941
+ "step": 133
942
+ },
943
+ {
944
+ "epoch": 0.5023430178069354,
945
+ "grad_norm": 2.7233939170837402,
946
+ "learning_rate": 8.959683216057512e-06,
947
+ "loss": 0.8723,
948
+ "step": 134
949
+ },
950
+ {
951
+ "epoch": 0.5060918462980318,
952
+ "grad_norm": 3.2135050296783447,
953
+ "learning_rate": 8.940689501724737e-06,
954
+ "loss": 0.9622,
955
+ "step": 135
956
+ },
957
+ {
958
+ "epoch": 0.5098406747891284,
959
+ "grad_norm": 2.69746994972229,
960
+ "learning_rate": 8.921544481913218e-06,
961
+ "loss": 0.9274,
962
+ "step": 136
963
+ },
964
+ {
965
+ "epoch": 0.5135895032802249,
966
+ "grad_norm": 2.718744993209839,
967
+ "learning_rate": 8.902248891709133e-06,
968
+ "loss": 0.7959,
969
+ "step": 137
970
+ },
971
+ {
972
+ "epoch": 0.5173383317713215,
973
+ "grad_norm": 2.5741381645202637,
974
+ "learning_rate": 8.882803471979917e-06,
975
+ "loss": 0.8901,
976
+ "step": 138
977
+ },
978
+ {
979
+ "epoch": 0.521087160262418,
980
+ "grad_norm": 2.678241491317749,
981
+ "learning_rate": 8.86320896934581e-06,
982
+ "loss": 0.8386,
983
+ "step": 139
984
+ },
985
+ {
986
+ "epoch": 0.5248359887535146,
987
+ "grad_norm": 3.3864355087280273,
988
+ "learning_rate": 8.843466136151191e-06,
989
+ "loss": 0.8401,
990
+ "step": 140
991
+ },
992
+ {
993
+ "epoch": 0.528584817244611,
994
+ "grad_norm": 2.4285075664520264,
995
+ "learning_rate": 8.823575730435694e-06,
996
+ "loss": 0.7686,
997
+ "step": 141
998
+ },
999
+ {
1000
+ "epoch": 0.5323336457357076,
1001
+ "grad_norm": 2.551071882247925,
1002
+ "learning_rate": 8.803538515905102e-06,
1003
+ "loss": 0.7005,
1004
+ "step": 142
1005
+ },
1006
+ {
1007
+ "epoch": 0.5360824742268041,
1008
+ "grad_norm": 2.8349006175994873,
1009
+ "learning_rate": 8.783355261902023e-06,
1010
+ "loss": 0.7764,
1011
+ "step": 143
1012
+ },
1013
+ {
1014
+ "epoch": 0.5398313027179007,
1015
+ "grad_norm": 2.5966458320617676,
1016
+ "learning_rate": 8.763026743376349e-06,
1017
+ "loss": 0.6911,
1018
+ "step": 144
1019
+ },
1020
+ {
1021
+ "epoch": 0.5435801312089972,
1022
+ "grad_norm": 2.6098828315734863,
1023
+ "learning_rate": 8.742553740855507e-06,
1024
+ "loss": 0.8884,
1025
+ "step": 145
1026
+ },
1027
+ {
1028
+ "epoch": 0.5473289597000938,
1029
+ "grad_norm": 2.8855576515197754,
1030
+ "learning_rate": 8.721937040414481e-06,
1031
+ "loss": 0.7172,
1032
+ "step": 146
1033
+ },
1034
+ {
1035
+ "epoch": 0.5510777881911902,
1036
+ "grad_norm": 3.230717420578003,
1037
+ "learning_rate": 8.70117743364564e-06,
1038
+ "loss": 0.7861,
1039
+ "step": 147
1040
+ },
1041
+ {
1042
+ "epoch": 0.5548266166822868,
1043
+ "grad_norm": 2.7732186317443848,
1044
+ "learning_rate": 8.680275717628336e-06,
1045
+ "loss": 0.657,
1046
+ "step": 148
1047
+ },
1048
+ {
1049
+ "epoch": 0.5585754451733833,
1050
+ "grad_norm": 2.7117972373962402,
1051
+ "learning_rate": 8.659232694898307e-06,
1052
+ "loss": 0.8825,
1053
+ "step": 149
1054
+ },
1055
+ {
1056
+ "epoch": 0.5623242736644799,
1057
+ "grad_norm": 2.5794482231140137,
1058
+ "learning_rate": 8.638049173416855e-06,
1059
+ "loss": 0.8581,
1060
+ "step": 150
1061
+ },
1062
+ {
1063
+ "epoch": 0.5660731021555764,
1064
+ "grad_norm": 2.9475417137145996,
1065
+ "learning_rate": 8.616725966539831e-06,
1066
+ "loss": 0.908,
1067
+ "step": 151
1068
+ },
1069
+ {
1070
+ "epoch": 0.569821930646673,
1071
+ "grad_norm": 2.9310085773468018,
1072
+ "learning_rate": 8.595263892986403e-06,
1073
+ "loss": 0.986,
1074
+ "step": 152
1075
+ },
1076
+ {
1077
+ "epoch": 0.5735707591377694,
1078
+ "grad_norm": 2.635514974594116,
1079
+ "learning_rate": 8.573663776807615e-06,
1080
+ "loss": 0.6768,
1081
+ "step": 153
1082
+ },
1083
+ {
1084
+ "epoch": 0.5773195876288659,
1085
+ "grad_norm": 2.826488733291626,
1086
+ "learning_rate": 8.551926447354759e-06,
1087
+ "loss": 1.0033,
1088
+ "step": 154
1089
+ },
1090
+ {
1091
+ "epoch": 0.5810684161199625,
1092
+ "grad_norm": 2.487631320953369,
1093
+ "learning_rate": 8.530052739247522e-06,
1094
+ "loss": 0.7031,
1095
+ "step": 155
1096
+ },
1097
+ {
1098
+ "epoch": 0.584817244611059,
1099
+ "grad_norm": 2.5516085624694824,
1100
+ "learning_rate": 8.508043492341944e-06,
1101
+ "loss": 0.7245,
1102
+ "step": 156
1103
+ },
1104
+ {
1105
+ "epoch": 0.5885660731021556,
1106
+ "grad_norm": 2.3507580757141113,
1107
+ "learning_rate": 8.485899551698166e-06,
1108
+ "loss": 0.8073,
1109
+ "step": 157
1110
+ },
1111
+ {
1112
+ "epoch": 0.5923149015932521,
1113
+ "grad_norm": 2.7757110595703125,
1114
+ "learning_rate": 8.463621767547998e-06,
1115
+ "loss": 0.8046,
1116
+ "step": 158
1117
+ },
1118
+ {
1119
+ "epoch": 0.5960637300843487,
1120
+ "grad_norm": 2.639394521713257,
1121
+ "learning_rate": 8.44121099526225e-06,
1122
+ "loss": 0.7603,
1123
+ "step": 159
1124
+ },
1125
+ {
1126
+ "epoch": 0.5998125585754451,
1127
+ "grad_norm": 2.907921314239502,
1128
+ "learning_rate": 8.418668095317912e-06,
1129
+ "loss": 0.9165,
1130
+ "step": 160
1131
+ },
1132
+ {
1133
+ "epoch": 0.6035613870665417,
1134
+ "grad_norm": 2.7137627601623535,
1135
+ "learning_rate": 8.395993933265102e-06,
1136
+ "loss": 0.8256,
1137
+ "step": 161
1138
+ },
1139
+ {
1140
+ "epoch": 0.6073102155576382,
1141
+ "grad_norm": 2.4688782691955566,
1142
+ "learning_rate": 8.373189379693838e-06,
1143
+ "loss": 0.9909,
1144
+ "step": 162
1145
+ },
1146
+ {
1147
+ "epoch": 0.6110590440487348,
1148
+ "grad_norm": 2.68072772026062,
1149
+ "learning_rate": 8.350255310200611e-06,
1150
+ "loss": 0.9874,
1151
+ "step": 163
1152
+ },
1153
+ {
1154
+ "epoch": 0.6148078725398313,
1155
+ "grad_norm": 3.0509040355682373,
1156
+ "learning_rate": 8.327192605354766e-06,
1157
+ "loss": 0.9298,
1158
+ "step": 164
1159
+ },
1160
+ {
1161
+ "epoch": 0.6185567010309279,
1162
+ "grad_norm": 2.4764254093170166,
1163
+ "learning_rate": 8.304002150664684e-06,
1164
+ "loss": 0.7797,
1165
+ "step": 165
1166
+ },
1167
+ {
1168
+ "epoch": 0.6223055295220243,
1169
+ "grad_norm": 2.992629051208496,
1170
+ "learning_rate": 8.280684836543794e-06,
1171
+ "loss": 0.8159,
1172
+ "step": 166
1173
+ },
1174
+ {
1175
+ "epoch": 0.6260543580131209,
1176
+ "grad_norm": 2.652869462966919,
1177
+ "learning_rate": 8.257241558276381e-06,
1178
+ "loss": 0.597,
1179
+ "step": 167
1180
+ },
1181
+ {
1182
+ "epoch": 0.6298031865042174,
1183
+ "grad_norm": 2.40967059135437,
1184
+ "learning_rate": 8.233673215983207e-06,
1185
+ "loss": 0.7821,
1186
+ "step": 168
1187
+ },
1188
+ {
1189
+ "epoch": 0.633552014995314,
1190
+ "grad_norm": 2.9071333408355713,
1191
+ "learning_rate": 8.209980714586955e-06,
1192
+ "loss": 0.9107,
1193
+ "step": 169
1194
+ },
1195
+ {
1196
+ "epoch": 0.6373008434864105,
1197
+ "grad_norm": 2.4595487117767334,
1198
+ "learning_rate": 8.18616496377748e-06,
1199
+ "loss": 0.8695,
1200
+ "step": 170
1201
+ },
1202
+ {
1203
+ "epoch": 0.6410496719775071,
1204
+ "grad_norm": 2.700758934020996,
1205
+ "learning_rate": 8.162226877976886e-06,
1206
+ "loss": 0.8954,
1207
+ "step": 171
1208
+ },
1209
+ {
1210
+ "epoch": 0.6447985004686035,
1211
+ "grad_norm": 3.0076358318328857,
1212
+ "learning_rate": 8.138167376304411e-06,
1213
+ "loss": 0.7885,
1214
+ "step": 172
1215
+ },
1216
+ {
1217
+ "epoch": 0.6485473289597001,
1218
+ "grad_norm": 2.321608543395996,
1219
+ "learning_rate": 8.113987382541138e-06,
1220
+ "loss": 0.6624,
1221
+ "step": 173
1222
+ },
1223
+ {
1224
+ "epoch": 0.6522961574507966,
1225
+ "grad_norm": 2.736940383911133,
1226
+ "learning_rate": 8.089687825094524e-06,
1227
+ "loss": 0.9155,
1228
+ "step": 174
1229
+ },
1230
+ {
1231
+ "epoch": 0.6560449859418932,
1232
+ "grad_norm": 2.9796676635742188,
1233
+ "learning_rate": 8.065269636962765e-06,
1234
+ "loss": 0.8578,
1235
+ "step": 175
1236
+ },
1237
+ {
1238
+ "epoch": 0.6597938144329897,
1239
+ "grad_norm": 2.3743815422058105,
1240
+ "learning_rate": 8.040733755698954e-06,
1241
+ "loss": 0.831,
1242
+ "step": 176
1243
+ },
1244
+ {
1245
+ "epoch": 0.6635426429240863,
1246
+ "grad_norm": 2.536311388015747,
1247
+ "learning_rate": 8.016081123375098e-06,
1248
+ "loss": 0.6518,
1249
+ "step": 177
1250
+ },
1251
+ {
1252
+ "epoch": 0.6672914714151827,
1253
+ "grad_norm": 2.936393976211548,
1254
+ "learning_rate": 7.991312686545939e-06,
1255
+ "loss": 0.8743,
1256
+ "step": 178
1257
+ },
1258
+ {
1259
+ "epoch": 0.6710402999062793,
1260
+ "grad_norm": 2.9552223682403564,
1261
+ "learning_rate": 7.96642939621261e-06,
1262
+ "loss": 1.0273,
1263
+ "step": 179
1264
+ },
1265
+ {
1266
+ "epoch": 0.6747891283973758,
1267
+ "grad_norm": 2.538328170776367,
1268
+ "learning_rate": 7.94143220778613e-06,
1269
+ "loss": 0.786,
1270
+ "step": 180
1271
+ },
1272
+ {
1273
+ "epoch": 0.6785379568884724,
1274
+ "grad_norm": 2.406303882598877,
1275
+ "learning_rate": 7.916322081050708e-06,
1276
+ "loss": 0.8003,
1277
+ "step": 181
1278
+ },
1279
+ {
1280
+ "epoch": 0.6822867853795689,
1281
+ "grad_norm": 2.377192735671997,
1282
+ "learning_rate": 7.8910999801269e-06,
1283
+ "loss": 0.7355,
1284
+ "step": 182
1285
+ },
1286
+ {
1287
+ "epoch": 0.6860356138706654,
1288
+ "grad_norm": 3.0302469730377197,
1289
+ "learning_rate": 7.865766873434582e-06,
1290
+ "loss": 1.0152,
1291
+ "step": 183
1292
+ },
1293
+ {
1294
+ "epoch": 0.689784442361762,
1295
+ "grad_norm": 2.5823426246643066,
1296
+ "learning_rate": 7.84032373365578e-06,
1297
+ "loss": 0.8919,
1298
+ "step": 184
1299
+ },
1300
+ {
1301
+ "epoch": 0.6935332708528584,
1302
+ "grad_norm": 2.891167640686035,
1303
+ "learning_rate": 7.814771537697312e-06,
1304
+ "loss": 0.8386,
1305
+ "step": 185
1306
+ },
1307
+ {
1308
+ "epoch": 0.697282099343955,
1309
+ "grad_norm": 2.61259126663208,
1310
+ "learning_rate": 7.789111266653285e-06,
1311
+ "loss": 0.8342,
1312
+ "step": 186
1313
+ },
1314
+ {
1315
+ "epoch": 0.7010309278350515,
1316
+ "grad_norm": 3.1016290187835693,
1317
+ "learning_rate": 7.76334390576742e-06,
1318
+ "loss": 0.8578,
1319
+ "step": 187
1320
+ },
1321
+ {
1322
+ "epoch": 0.7047797563261481,
1323
+ "grad_norm": 2.2901554107666016,
1324
+ "learning_rate": 7.737470444395227e-06,
1325
+ "loss": 0.6153,
1326
+ "step": 188
1327
+ },
1328
+ {
1329
+ "epoch": 0.7085285848172446,
1330
+ "grad_norm": 3.0746381282806396,
1331
+ "learning_rate": 7.71149187596602e-06,
1332
+ "loss": 0.7487,
1333
+ "step": 189
1334
+ },
1335
+ {
1336
+ "epoch": 0.7122774133083412,
1337
+ "grad_norm": 2.534573554992676,
1338
+ "learning_rate": 7.685409197944768e-06,
1339
+ "loss": 0.9354,
1340
+ "step": 190
1341
+ },
1342
+ {
1343
+ "epoch": 0.7160262417994376,
1344
+ "grad_norm": 2.7725582122802734,
1345
+ "learning_rate": 7.6592234117938e-06,
1346
+ "loss": 0.8641,
1347
+ "step": 191
1348
+ },
1349
+ {
1350
+ "epoch": 0.7197750702905342,
1351
+ "grad_norm": 2.284388780593872,
1352
+ "learning_rate": 7.63293552293435e-06,
1353
+ "loss": 0.8296,
1354
+ "step": 192
1355
+ },
1356
+ {
1357
+ "epoch": 0.7235238987816307,
1358
+ "grad_norm": 2.665536880493164,
1359
+ "learning_rate": 7.60654654070796e-06,
1360
+ "loss": 0.7976,
1361
+ "step": 193
1362
+ },
1363
+ {
1364
+ "epoch": 0.7272727272727273,
1365
+ "grad_norm": 2.496126651763916,
1366
+ "learning_rate": 7.580057478337717e-06,
1367
+ "loss": 0.8097,
1368
+ "step": 194
1369
+ },
1370
+ {
1371
+ "epoch": 0.7310215557638238,
1372
+ "grad_norm": 2.898500919342041,
1373
+ "learning_rate": 7.553469352889356e-06,
1374
+ "loss": 0.832,
1375
+ "step": 195
1376
+ },
1377
+ {
1378
+ "epoch": 0.7347703842549204,
1379
+ "grad_norm": 2.545947551727295,
1380
+ "learning_rate": 7.526783185232208e-06,
1381
+ "loss": 0.8465,
1382
+ "step": 196
1383
+ },
1384
+ {
1385
+ "epoch": 0.7385192127460168,
1386
+ "grad_norm": 2.320594549179077,
1387
+ "learning_rate": 7.500000000000001e-06,
1388
+ "loss": 0.6391,
1389
+ "step": 197
1390
+ },
1391
+ {
1392
+ "epoch": 0.7422680412371134,
1393
+ "grad_norm": 2.390089511871338,
1394
+ "learning_rate": 7.473120825551517e-06,
1395
+ "loss": 0.8494,
1396
+ "step": 198
1397
+ },
1398
+ {
1399
+ "epoch": 0.7460168697282099,
1400
+ "grad_norm": 3.2954750061035156,
1401
+ "learning_rate": 7.446146693931111e-06,
1402
+ "loss": 0.9194,
1403
+ "step": 199
1404
+ },
1405
+ {
1406
+ "epoch": 0.7497656982193065,
1407
+ "grad_norm": 3.3448970317840576,
1408
+ "learning_rate": 7.419078640829088e-06,
1409
+ "loss": 0.8001,
1410
+ "step": 200
1411
+ },
1412
+ {
1413
+ "epoch": 0.753514526710403,
1414
+ "grad_norm": 2.490509033203125,
1415
+ "learning_rate": 7.391917705541927e-06,
1416
+ "loss": 0.8049,
1417
+ "step": 201
1418
+ },
1419
+ {
1420
+ "epoch": 0.7572633552014996,
1421
+ "grad_norm": 3.141767740249634,
1422
+ "learning_rate": 7.364664930932385e-06,
1423
+ "loss": 0.6791,
1424
+ "step": 202
1425
+ },
1426
+ {
1427
+ "epoch": 0.761012183692596,
1428
+ "grad_norm": 2.646108388900757,
1429
+ "learning_rate": 7.337321363389453e-06,
1430
+ "loss": 0.6114,
1431
+ "step": 203
1432
+ },
1433
+ {
1434
+ "epoch": 0.7647610121836926,
1435
+ "grad_norm": 2.71254563331604,
1436
+ "learning_rate": 7.3098880527881755e-06,
1437
+ "loss": 0.8919,
1438
+ "step": 204
1439
+ },
1440
+ {
1441
+ "epoch": 0.7685098406747891,
1442
+ "grad_norm": 2.701137065887451,
1443
+ "learning_rate": 7.282366052449351e-06,
1444
+ "loss": 0.9196,
1445
+ "step": 205
1446
+ },
1447
+ {
1448
+ "epoch": 0.7722586691658857,
1449
+ "grad_norm": 2.5790905952453613,
1450
+ "learning_rate": 7.254756419099074e-06,
1451
+ "loss": 0.8328,
1452
+ "step": 206
1453
+ },
1454
+ {
1455
+ "epoch": 0.7760074976569822,
1456
+ "grad_norm": 2.885322332382202,
1457
+ "learning_rate": 7.227060212828171e-06,
1458
+ "loss": 0.7947,
1459
+ "step": 207
1460
+ },
1461
+ {
1462
+ "epoch": 0.7797563261480788,
1463
+ "grad_norm": 2.563446044921875,
1464
+ "learning_rate": 7.199278497051498e-06,
1465
+ "loss": 0.8801,
1466
+ "step": 208
1467
+ },
1468
+ {
1469
+ "epoch": 0.7835051546391752,
1470
+ "grad_norm": 2.7891945838928223,
1471
+ "learning_rate": 7.171412338467101e-06,
1472
+ "loss": 0.7966,
1473
+ "step": 209
1474
+ },
1475
+ {
1476
+ "epoch": 0.7872539831302718,
1477
+ "grad_norm": 2.888983726501465,
1478
+ "learning_rate": 7.143462807015271e-06,
1479
+ "loss": 0.7454,
1480
+ "step": 210
1481
+ },
1482
+ {
1483
+ "epoch": 0.7910028116213683,
1484
+ "grad_norm": 2.7022621631622314,
1485
+ "learning_rate": 7.115430975837457e-06,
1486
+ "loss": 0.9606,
1487
+ "step": 211
1488
+ },
1489
+ {
1490
+ "epoch": 0.7947516401124649,
1491
+ "grad_norm": 2.689760208129883,
1492
+ "learning_rate": 7.087317921235059e-06,
1493
+ "loss": 0.819,
1494
+ "step": 212
1495
+ },
1496
+ {
1497
+ "epoch": 0.7985004686035614,
1498
+ "grad_norm": 2.960847854614258,
1499
+ "learning_rate": 7.059124722628113e-06,
1500
+ "loss": 0.9319,
1501
+ "step": 213
1502
+ },
1503
+ {
1504
+ "epoch": 0.8022492970946579,
1505
+ "grad_norm": 3.116780996322632,
1506
+ "learning_rate": 7.030852462513827e-06,
1507
+ "loss": 1.0187,
1508
+ "step": 214
1509
+ },
1510
+ {
1511
+ "epoch": 0.8059981255857545,
1512
+ "grad_norm": 2.5313186645507812,
1513
+ "learning_rate": 7.002502226425042e-06,
1514
+ "loss": 0.9004,
1515
+ "step": 215
1516
+ },
1517
+ {
1518
+ "epoch": 0.8097469540768509,
1519
+ "grad_norm": 2.4676010608673096,
1520
+ "learning_rate": 6.974075102888535e-06,
1521
+ "loss": 0.805,
1522
+ "step": 216
1523
+ },
1524
+ {
1525
+ "epoch": 0.8134957825679475,
1526
+ "grad_norm": 2.8545048236846924,
1527
+ "learning_rate": 6.945572183383229e-06,
1528
+ "loss": 0.8092,
1529
+ "step": 217
1530
+ },
1531
+ {
1532
+ "epoch": 0.817244611059044,
1533
+ "grad_norm": 2.710291624069214,
1534
+ "learning_rate": 6.916994562298286e-06,
1535
+ "loss": 0.8639,
1536
+ "step": 218
1537
+ },
1538
+ {
1539
+ "epoch": 0.8209934395501406,
1540
+ "grad_norm": 2.850886344909668,
1541
+ "learning_rate": 6.888343336891088e-06,
1542
+ "loss": 0.8063,
1543
+ "step": 219
1544
+ },
1545
+ {
1546
+ "epoch": 0.8247422680412371,
1547
+ "grad_norm": 2.6347622871398926,
1548
+ "learning_rate": 6.859619607245102e-06,
1549
+ "loss": 0.9418,
1550
+ "step": 220
1551
+ },
1552
+ {
1553
+ "epoch": 0.8284910965323337,
1554
+ "grad_norm": 2.4911224842071533,
1555
+ "learning_rate": 6.830824476227646e-06,
1556
+ "loss": 0.633,
1557
+ "step": 221
1558
+ },
1559
+ {
1560
+ "epoch": 0.8322399250234301,
1561
+ "grad_norm": 2.8799281120300293,
1562
+ "learning_rate": 6.801959049447546e-06,
1563
+ "loss": 0.7747,
1564
+ "step": 222
1565
+ },
1566
+ {
1567
+ "epoch": 0.8359887535145267,
1568
+ "grad_norm": 2.7227516174316406,
1569
+ "learning_rate": 6.773024435212678e-06,
1570
+ "loss": 1.0301,
1571
+ "step": 223
1572
+ },
1573
+ {
1574
+ "epoch": 0.8397375820056232,
1575
+ "grad_norm": 2.352261781692505,
1576
+ "learning_rate": 6.744021744487422e-06,
1577
+ "loss": 0.7293,
1578
+ "step": 224
1579
+ },
1580
+ {
1581
+ "epoch": 0.8434864104967198,
1582
+ "grad_norm": 2.8626327514648438,
1583
+ "learning_rate": 6.714952090849996e-06,
1584
+ "loss": 1.022,
1585
+ "step": 225
1586
+ },
1587
+ {
1588
+ "epoch": 0.8472352389878163,
1589
+ "grad_norm": 2.3415591716766357,
1590
+ "learning_rate": 6.685816590449708e-06,
1591
+ "loss": 0.7533,
1592
+ "step": 226
1593
+ },
1594
+ {
1595
+ "epoch": 0.8509840674789129,
1596
+ "grad_norm": 2.774419069290161,
1597
+ "learning_rate": 6.6566163619641e-06,
1598
+ "loss": 0.7825,
1599
+ "step": 227
1600
+ },
1601
+ {
1602
+ "epoch": 0.8547328959700093,
1603
+ "grad_norm": 2.38519287109375,
1604
+ "learning_rate": 6.62735252655599e-06,
1605
+ "loss": 0.9026,
1606
+ "step": 228
1607
+ },
1608
+ {
1609
+ "epoch": 0.8584817244611059,
1610
+ "grad_norm": 2.562180280685425,
1611
+ "learning_rate": 6.598026207830428e-06,
1612
+ "loss": 0.8274,
1613
+ "step": 229
1614
+ },
1615
+ {
1616
+ "epoch": 0.8622305529522024,
1617
+ "grad_norm": 2.506265640258789,
1618
+ "learning_rate": 6.568638531791555e-06,
1619
+ "loss": 0.8533,
1620
+ "step": 230
1621
+ },
1622
+ {
1623
+ "epoch": 0.865979381443299,
1624
+ "grad_norm": 2.4109628200531006,
1625
+ "learning_rate": 6.539190626799366e-06,
1626
+ "loss": 0.718,
1627
+ "step": 231
1628
+ },
1629
+ {
1630
+ "epoch": 0.8697282099343955,
1631
+ "grad_norm": 2.856254816055298,
1632
+ "learning_rate": 6.5096836235263904e-06,
1633
+ "loss": 0.8667,
1634
+ "step": 232
1635
+ },
1636
+ {
1637
+ "epoch": 0.8734770384254921,
1638
+ "grad_norm": 2.8561127185821533,
1639
+ "learning_rate": 6.480118654914276e-06,
1640
+ "loss": 0.5914,
1641
+ "step": 233
1642
+ },
1643
+ {
1644
+ "epoch": 0.8772258669165885,
1645
+ "grad_norm": 2.726389169692993,
1646
+ "learning_rate": 6.4504968561302905e-06,
1647
+ "loss": 0.9917,
1648
+ "step": 234
1649
+ },
1650
+ {
1651
+ "epoch": 0.8809746954076851,
1652
+ "grad_norm": 2.7317206859588623,
1653
+ "learning_rate": 6.4208193645237314e-06,
1654
+ "loss": 0.9438,
1655
+ "step": 235
1656
+ },
1657
+ {
1658
+ "epoch": 0.8847235238987816,
1659
+ "grad_norm": 2.6856586933135986,
1660
+ "learning_rate": 6.391087319582264e-06,
1661
+ "loss": 0.8312,
1662
+ "step": 236
1663
+ },
1664
+ {
1665
+ "epoch": 0.8884723523898782,
1666
+ "grad_norm": 2.7664246559143066,
1667
+ "learning_rate": 6.3613018628881655e-06,
1668
+ "loss": 0.8756,
1669
+ "step": 237
1670
+ },
1671
+ {
1672
+ "epoch": 0.8922211808809747,
1673
+ "grad_norm": 2.266479730606079,
1674
+ "learning_rate": 6.331464138074493e-06,
1675
+ "loss": 0.7957,
1676
+ "step": 238
1677
+ },
1678
+ {
1679
+ "epoch": 0.8959700093720713,
1680
+ "grad_norm": 2.8850481510162354,
1681
+ "learning_rate": 6.301575290781174e-06,
1682
+ "loss": 0.9892,
1683
+ "step": 239
1684
+ },
1685
+ {
1686
+ "epoch": 0.8997188378631678,
1687
+ "grad_norm": 3.300827980041504,
1688
+ "learning_rate": 6.271636468611022e-06,
1689
+ "loss": 0.8914,
1690
+ "step": 240
1691
+ },
1692
+ {
1693
+ "epoch": 0.9034676663542643,
1694
+ "grad_norm": 2.5576953887939453,
1695
+ "learning_rate": 6.241648821085666e-06,
1696
+ "loss": 0.9061,
1697
+ "step": 241
1698
+ },
1699
+ {
1700
+ "epoch": 0.9072164948453608,
1701
+ "grad_norm": 2.373811721801758,
1702
+ "learning_rate": 6.211613499601419e-06,
1703
+ "loss": 0.8346,
1704
+ "step": 242
1705
+ },
1706
+ {
1707
+ "epoch": 0.9109653233364574,
1708
+ "grad_norm": 2.5030999183654785,
1709
+ "learning_rate": 6.181531657385068e-06,
1710
+ "loss": 0.8327,
1711
+ "step": 243
1712
+ },
1713
+ {
1714
+ "epoch": 0.9147141518275539,
1715
+ "grad_norm": 3.0258920192718506,
1716
+ "learning_rate": 6.1514044494496e-06,
1717
+ "loss": 0.7557,
1718
+ "step": 244
1719
+ },
1720
+ {
1721
+ "epoch": 0.9184629803186504,
1722
+ "grad_norm": 2.5062878131866455,
1723
+ "learning_rate": 6.1212330325498425e-06,
1724
+ "loss": 0.8887,
1725
+ "step": 245
1726
+ },
1727
+ {
1728
+ "epoch": 0.922211808809747,
1729
+ "grad_norm": 2.974062442779541,
1730
+ "learning_rate": 6.091018565138062e-06,
1731
+ "loss": 0.8752,
1732
+ "step": 246
1733
+ },
1734
+ {
1735
+ "epoch": 0.9259606373008434,
1736
+ "grad_norm": 2.3620338439941406,
1737
+ "learning_rate": 6.060762207319479e-06,
1738
+ "loss": 0.5664,
1739
+ "step": 247
1740
+ },
1741
+ {
1742
+ "epoch": 0.92970946579194,
1743
+ "grad_norm": 2.6449522972106934,
1744
+ "learning_rate": 6.03046512080772e-06,
1745
+ "loss": 0.7479,
1746
+ "step": 248
1747
+ },
1748
+ {
1749
+ "epoch": 0.9334582942830365,
1750
+ "grad_norm": 2.6556012630462646,
1751
+ "learning_rate": 6.000128468880223e-06,
1752
+ "loss": 0.9128,
1753
+ "step": 249
1754
+ },
1755
+ {
1756
+ "epoch": 0.9372071227741331,
1757
+ "grad_norm": 2.5893819332122803,
1758
+ "learning_rate": 5.9697534163335645e-06,
1759
+ "loss": 0.7899,
1760
+ "step": 250
1761
+ },
1762
+ {
1763
+ "epoch": 0.9409559512652296,
1764
+ "grad_norm": 2.8196489810943604,
1765
+ "learning_rate": 5.939341129438739e-06,
1766
+ "loss": 0.9707,
1767
+ "step": 251
1768
+ },
1769
+ {
1770
+ "epoch": 0.9447047797563262,
1771
+ "grad_norm": 2.6079866886138916,
1772
+ "learning_rate": 5.908892775896383e-06,
1773
+ "loss": 0.6921,
1774
+ "step": 252
1775
+ },
1776
+ {
1777
+ "epoch": 0.9484536082474226,
1778
+ "grad_norm": 2.749600648880005,
1779
+ "learning_rate": 5.878409524791931e-06,
1780
+ "loss": 0.8562,
1781
+ "step": 253
1782
+ },
1783
+ {
1784
+ "epoch": 0.9522024367385192,
1785
+ "grad_norm": 2.7583858966827393,
1786
+ "learning_rate": 5.847892546550738e-06,
1787
+ "loss": 0.7389,
1788
+ "step": 254
1789
+ },
1790
+ {
1791
+ "epoch": 0.9559512652296157,
1792
+ "grad_norm": 2.819322347640991,
1793
+ "learning_rate": 5.817343012893132e-06,
1794
+ "loss": 0.9122,
1795
+ "step": 255
1796
+ },
1797
+ {
1798
+ "epoch": 0.9597000937207123,
1799
+ "grad_norm": 2.7966537475585938,
1800
+ "learning_rate": 5.786762096789431e-06,
1801
+ "loss": 1.0095,
1802
+ "step": 256
1803
+ },
1804
+ {
1805
+ "epoch": 0.9634489222118088,
1806
+ "grad_norm": 2.982713460922241,
1807
+ "learning_rate": 5.756150972414904e-06,
1808
+ "loss": 1.0745,
1809
+ "step": 257
1810
+ },
1811
+ {
1812
+ "epoch": 0.9671977507029054,
1813
+ "grad_norm": 2.7742886543273926,
1814
+ "learning_rate": 5.725510815104685e-06,
1815
+ "loss": 0.7517,
1816
+ "step": 258
1817
+ },
1818
+ {
1819
+ "epoch": 0.9709465791940018,
1820
+ "grad_norm": 2.983306646347046,
1821
+ "learning_rate": 5.694842801308651e-06,
1822
+ "loss": 0.8668,
1823
+ "step": 259
1824
+ },
1825
+ {
1826
+ "epoch": 0.9746954076850984,
1827
+ "grad_norm": 2.550466299057007,
1828
+ "learning_rate": 5.664148108546242e-06,
1829
+ "loss": 0.7585,
1830
+ "step": 260
1831
+ },
1832
+ {
1833
+ "epoch": 0.9784442361761949,
1834
+ "grad_norm": 2.5021555423736572,
1835
+ "learning_rate": 5.633427915361261e-06,
1836
+ "loss": 0.7487,
1837
+ "step": 261
1838
+ },
1839
+ {
1840
+ "epoch": 0.9821930646672915,
1841
+ "grad_norm": 2.9954419136047363,
1842
+ "learning_rate": 5.6026834012766155e-06,
1843
+ "loss": 0.7231,
1844
+ "step": 262
1845
+ },
1846
+ {
1847
+ "epoch": 0.985941893158388,
1848
+ "grad_norm": 2.553372859954834,
1849
+ "learning_rate": 5.5719157467490305e-06,
1850
+ "loss": 0.7305,
1851
+ "step": 263
1852
+ },
1853
+ {
1854
+ "epoch": 0.9896907216494846,
1855
+ "grad_norm": 2.377293825149536,
1856
+ "learning_rate": 5.541126133123721e-06,
1857
+ "loss": 0.9153,
1858
+ "step": 264
1859
+ },
1860
+ {
1861
+ "epoch": 0.993439550140581,
1862
+ "grad_norm": 2.6041624546051025,
1863
+ "learning_rate": 5.510315742589042e-06,
1864
+ "loss": 0.7345,
1865
+ "step": 265
1866
+ },
1867
+ {
1868
+ "epoch": 0.9971883786316776,
1869
+ "grad_norm": 2.1862921714782715,
1870
+ "learning_rate": 5.479485758131089e-06,
1871
+ "loss": 0.8135,
1872
+ "step": 266
1873
+ },
1874
+ {
1875
+ "epoch": 1.0,
1876
+ "grad_norm": 3.3813092708587646,
1877
+ "learning_rate": 5.4486373634882805e-06,
1878
+ "loss": 0.7056,
1879
+ "step": 267
1880
+ },
1881
+ {
1882
+ "epoch": 1.0037488284910965,
1883
+ "grad_norm": 2.4991369247436523,
1884
+ "learning_rate": 5.417771743105908e-06,
1885
+ "loss": 0.5993,
1886
+ "step": 268
1887
+ },
1888
+ {
1889
+ "epoch": 1.007497656982193,
1890
+ "grad_norm": 2.8168036937713623,
1891
+ "learning_rate": 5.386890082090652e-06,
1892
+ "loss": 0.6456,
1893
+ "step": 269
1894
+ },
1895
+ {
1896
+ "epoch": 1.0112464854732897,
1897
+ "grad_norm": 2.7016103267669678,
1898
+ "learning_rate": 5.355993566165091e-06,
1899
+ "loss": 0.7714,
1900
+ "step": 270
1901
+ },
1902
+ {
1903
+ "epoch": 1.0149953139643861,
1904
+ "grad_norm": 2.449699878692627,
1905
+ "learning_rate": 5.325083381622165e-06,
1906
+ "loss": 0.7332,
1907
+ "step": 271
1908
+ },
1909
+ {
1910
+ "epoch": 1.0187441424554826,
1911
+ "grad_norm": 2.5395665168762207,
1912
+ "learning_rate": 5.294160715279626e-06,
1913
+ "loss": 0.7395,
1914
+ "step": 272
1915
+ },
1916
+ {
1917
+ "epoch": 1.022492970946579,
1918
+ "grad_norm": 2.3110768795013428,
1919
+ "learning_rate": 5.263226754434481e-06,
1920
+ "loss": 0.6565,
1921
+ "step": 273
1922
+ },
1923
+ {
1924
+ "epoch": 1.0262417994376758,
1925
+ "grad_norm": 2.3739070892333984,
1926
+ "learning_rate": 5.232282686817392e-06,
1927
+ "loss": 0.7151,
1928
+ "step": 274
1929
+ },
1930
+ {
1931
+ "epoch": 1.0299906279287723,
1932
+ "grad_norm": 2.500891923904419,
1933
+ "learning_rate": 5.201329700547077e-06,
1934
+ "loss": 0.6505,
1935
+ "step": 275
1936
+ },
1937
+ {
1938
+ "epoch": 1.0337394564198688,
1939
+ "grad_norm": 2.575612783432007,
1940
+ "learning_rate": 5.170368984084695e-06,
1941
+ "loss": 0.737,
1942
+ "step": 276
1943
+ },
1944
+ {
1945
+ "epoch": 1.0374882849109652,
1946
+ "grad_norm": 2.4420552253723145,
1947
+ "learning_rate": 5.139401726188208e-06,
1948
+ "loss": 0.5698,
1949
+ "step": 277
1950
+ },
1951
+ {
1952
+ "epoch": 1.041237113402062,
1953
+ "grad_norm": 2.4866912364959717,
1954
+ "learning_rate": 5.108429115866744e-06,
1955
+ "loss": 0.5371,
1956
+ "step": 278
1957
+ },
1958
+ {
1959
+ "epoch": 1.0449859418931584,
1960
+ "grad_norm": 2.7172107696533203,
1961
+ "learning_rate": 5.077452342334939e-06,
1962
+ "loss": 0.7068,
1963
+ "step": 279
1964
+ },
1965
+ {
1966
+ "epoch": 1.0487347703842549,
1967
+ "grad_norm": 2.302205801010132,
1968
+ "learning_rate": 5.046472594967279e-06,
1969
+ "loss": 0.6946,
1970
+ "step": 280
1971
+ },
1972
+ {
1973
+ "epoch": 1.0524835988753514,
1974
+ "grad_norm": 2.5750224590301514,
1975
+ "learning_rate": 5.01549106325243e-06,
1976
+ "loss": 0.7505,
1977
+ "step": 281
1978
+ },
1979
+ {
1980
+ "epoch": 1.056232427366448,
1981
+ "grad_norm": 2.5373618602752686,
1982
+ "learning_rate": 4.9845089367475715e-06,
1983
+ "loss": 0.7652,
1984
+ "step": 282
1985
+ },
1986
+ {
1987
+ "epoch": 1.0599812558575445,
1988
+ "grad_norm": 2.721522092819214,
1989
+ "learning_rate": 4.953527405032723e-06,
1990
+ "loss": 0.7254,
1991
+ "step": 283
1992
+ },
1993
+ {
1994
+ "epoch": 1.063730084348641,
1995
+ "grad_norm": 2.75787091255188,
1996
+ "learning_rate": 4.922547657665062e-06,
1997
+ "loss": 0.8159,
1998
+ "step": 284
1999
+ },
2000
+ {
2001
+ "epoch": 1.0674789128397375,
2002
+ "grad_norm": 2.2756996154785156,
2003
+ "learning_rate": 4.891570884133256e-06,
2004
+ "loss": 0.5339,
2005
+ "step": 285
2006
+ },
2007
+ {
2008
+ "epoch": 1.0712277413308342,
2009
+ "grad_norm": 2.541193962097168,
2010
+ "learning_rate": 4.860598273811793e-06,
2011
+ "loss": 0.6664,
2012
+ "step": 286
2013
+ },
2014
+ {
2015
+ "epoch": 1.0749765698219307,
2016
+ "grad_norm": 2.9506993293762207,
2017
+ "learning_rate": 4.829631015915306e-06,
2018
+ "loss": 0.6293,
2019
+ "step": 287
2020
+ },
2021
+ {
2022
+ "epoch": 1.0787253983130272,
2023
+ "grad_norm": 2.649301290512085,
2024
+ "learning_rate": 4.798670299452926e-06,
2025
+ "loss": 0.5073,
2026
+ "step": 288
2027
+ },
2028
+ {
2029
+ "epoch": 1.0824742268041236,
2030
+ "grad_norm": 2.1914241313934326,
2031
+ "learning_rate": 4.767717313182611e-06,
2032
+ "loss": 0.6079,
2033
+ "step": 289
2034
+ },
2035
+ {
2036
+ "epoch": 1.0862230552952203,
2037
+ "grad_norm": 2.44549298286438,
2038
+ "learning_rate": 4.736773245565521e-06,
2039
+ "loss": 0.5781,
2040
+ "step": 290
2041
+ },
2042
+ {
2043
+ "epoch": 1.0899718837863168,
2044
+ "grad_norm": 2.684919595718384,
2045
+ "learning_rate": 4.705839284720376e-06,
2046
+ "loss": 0.5453,
2047
+ "step": 291
2048
+ },
2049
+ {
2050
+ "epoch": 1.0937207122774133,
2051
+ "grad_norm": 2.83229660987854,
2052
+ "learning_rate": 4.6749166183778375e-06,
2053
+ "loss": 0.831,
2054
+ "step": 292
2055
+ },
2056
+ {
2057
+ "epoch": 1.0974695407685098,
2058
+ "grad_norm": 2.6568431854248047,
2059
+ "learning_rate": 4.64400643383491e-06,
2060
+ "loss": 0.587,
2061
+ "step": 293
2062
+ },
2063
+ {
2064
+ "epoch": 1.1012183692596063,
2065
+ "grad_norm": 2.890794277191162,
2066
+ "learning_rate": 4.613109917909349e-06,
2067
+ "loss": 0.7931,
2068
+ "step": 294
2069
+ },
2070
+ {
2071
+ "epoch": 1.104967197750703,
2072
+ "grad_norm": 2.916185140609741,
2073
+ "learning_rate": 4.582228256894093e-06,
2074
+ "loss": 0.7106,
2075
+ "step": 295
2076
+ },
2077
+ {
2078
+ "epoch": 1.1087160262417994,
2079
+ "grad_norm": 2.470686912536621,
2080
+ "learning_rate": 4.55136263651172e-06,
2081
+ "loss": 0.7789,
2082
+ "step": 296
2083
+ },
2084
+ {
2085
+ "epoch": 1.112464854732896,
2086
+ "grad_norm": 2.4361371994018555,
2087
+ "learning_rate": 4.520514241868912e-06,
2088
+ "loss": 0.6496,
2089
+ "step": 297
2090
+ },
2091
+ {
2092
+ "epoch": 1.1162136832239926,
2093
+ "grad_norm": 2.7543270587921143,
2094
+ "learning_rate": 4.489684257410959e-06,
2095
+ "loss": 0.6137,
2096
+ "step": 298
2097
+ },
2098
+ {
2099
+ "epoch": 1.119962511715089,
2100
+ "grad_norm": 2.6973302364349365,
2101
+ "learning_rate": 4.458873866876282e-06,
2102
+ "loss": 0.5914,
2103
+ "step": 299
2104
+ },
2105
+ {
2106
+ "epoch": 1.1237113402061856,
2107
+ "grad_norm": 2.2894020080566406,
2108
+ "learning_rate": 4.428084253250972e-06,
2109
+ "loss": 0.66,
2110
+ "step": 300
2111
+ },
2112
+ {
2113
+ "epoch": 1.127460168697282,
2114
+ "grad_norm": 2.703667402267456,
2115
+ "learning_rate": 4.397316598723385e-06,
2116
+ "loss": 0.6326,
2117
+ "step": 301
2118
+ },
2119
+ {
2120
+ "epoch": 1.1312089971883785,
2121
+ "grad_norm": 2.5640034675598145,
2122
+ "learning_rate": 4.3665720846387406e-06,
2123
+ "loss": 0.6076,
2124
+ "step": 302
2125
+ },
2126
+ {
2127
+ "epoch": 1.1349578256794752,
2128
+ "grad_norm": 2.6277530193328857,
2129
+ "learning_rate": 4.335851891453759e-06,
2130
+ "loss": 0.6344,
2131
+ "step": 303
2132
+ },
2133
+ {
2134
+ "epoch": 1.1387066541705717,
2135
+ "grad_norm": 2.1854584217071533,
2136
+ "learning_rate": 4.305157198691351e-06,
2137
+ "loss": 0.537,
2138
+ "step": 304
2139
+ },
2140
+ {
2141
+ "epoch": 1.1424554826616682,
2142
+ "grad_norm": 2.584956407546997,
2143
+ "learning_rate": 4.2744891848953156e-06,
2144
+ "loss": 0.6874,
2145
+ "step": 305
2146
+ },
2147
+ {
2148
+ "epoch": 1.1462043111527647,
2149
+ "grad_norm": 3.1633033752441406,
2150
+ "learning_rate": 4.2438490275850965e-06,
2151
+ "loss": 0.776,
2152
+ "step": 306
2153
+ },
2154
+ {
2155
+ "epoch": 1.1499531396438614,
2156
+ "grad_norm": 2.372377634048462,
2157
+ "learning_rate": 4.2132379032105695e-06,
2158
+ "loss": 0.6193,
2159
+ "step": 307
2160
+ },
2161
+ {
2162
+ "epoch": 1.1537019681349578,
2163
+ "grad_norm": 2.6179869174957275,
2164
+ "learning_rate": 4.182656987106869e-06,
2165
+ "loss": 0.8479,
2166
+ "step": 308
2167
+ },
2168
+ {
2169
+ "epoch": 1.1574507966260543,
2170
+ "grad_norm": 2.4828739166259766,
2171
+ "learning_rate": 4.152107453449263e-06,
2172
+ "loss": 0.5892,
2173
+ "step": 309
2174
+ },
2175
+ {
2176
+ "epoch": 1.1611996251171508,
2177
+ "grad_norm": 2.4343421459198,
2178
+ "learning_rate": 4.121590475208071e-06,
2179
+ "loss": 0.5757,
2180
+ "step": 310
2181
+ },
2182
+ {
2183
+ "epoch": 1.1649484536082475,
2184
+ "grad_norm": 2.660548210144043,
2185
+ "learning_rate": 4.091107224103619e-06,
2186
+ "loss": 0.7259,
2187
+ "step": 311
2188
+ },
2189
+ {
2190
+ "epoch": 1.168697282099344,
2191
+ "grad_norm": 2.5796167850494385,
2192
+ "learning_rate": 4.060658870561263e-06,
2193
+ "loss": 0.5966,
2194
+ "step": 312
2195
+ },
2196
+ {
2197
+ "epoch": 1.1724461105904405,
2198
+ "grad_norm": 2.7497189044952393,
2199
+ "learning_rate": 4.030246583666437e-06,
2200
+ "loss": 0.6979,
2201
+ "step": 313
2202
+ },
2203
+ {
2204
+ "epoch": 1.176194939081537,
2205
+ "grad_norm": 2.582590103149414,
2206
+ "learning_rate": 3.999871531119779e-06,
2207
+ "loss": 0.722,
2208
+ "step": 314
2209
+ },
2210
+ {
2211
+ "epoch": 1.1799437675726336,
2212
+ "grad_norm": 2.6821389198303223,
2213
+ "learning_rate": 3.969534879192281e-06,
2214
+ "loss": 0.7015,
2215
+ "step": 315
2216
+ },
2217
+ {
2218
+ "epoch": 1.1836925960637301,
2219
+ "grad_norm": 2.4163970947265625,
2220
+ "learning_rate": 3.9392377926805226e-06,
2221
+ "loss": 0.4959,
2222
+ "step": 316
2223
+ },
2224
+ {
2225
+ "epoch": 1.1874414245548266,
2226
+ "grad_norm": 2.1838912963867188,
2227
+ "learning_rate": 3.9089814348619386e-06,
2228
+ "loss": 0.6188,
2229
+ "step": 317
2230
+ },
2231
+ {
2232
+ "epoch": 1.191190253045923,
2233
+ "grad_norm": 2.767521619796753,
2234
+ "learning_rate": 3.878766967450158e-06,
2235
+ "loss": 0.8107,
2236
+ "step": 318
2237
+ },
2238
+ {
2239
+ "epoch": 1.1949390815370198,
2240
+ "grad_norm": 2.630614995956421,
2241
+ "learning_rate": 3.848595550550401e-06,
2242
+ "loss": 0.7719,
2243
+ "step": 319
2244
+ },
2245
+ {
2246
+ "epoch": 1.1986879100281163,
2247
+ "grad_norm": 2.7656266689300537,
2248
+ "learning_rate": 3.818468342614932e-06,
2249
+ "loss": 0.6614,
2250
+ "step": 320
2251
+ },
2252
+ {
2253
+ "epoch": 1.2024367385192127,
2254
+ "grad_norm": 2.6769886016845703,
2255
+ "learning_rate": 3.788386500398583e-06,
2256
+ "loss": 0.6055,
2257
+ "step": 321
2258
+ },
2259
+ {
2260
+ "epoch": 1.2061855670103092,
2261
+ "grad_norm": 2.4502344131469727,
2262
+ "learning_rate": 3.758351178914336e-06,
2263
+ "loss": 0.5417,
2264
+ "step": 322
2265
+ },
2266
+ {
2267
+ "epoch": 1.209934395501406,
2268
+ "grad_norm": 2.5828676223754883,
2269
+ "learning_rate": 3.728363531388979e-06,
2270
+ "loss": 0.6551,
2271
+ "step": 323
2272
+ },
2273
+ {
2274
+ "epoch": 1.2136832239925024,
2275
+ "grad_norm": 2.5377285480499268,
2276
+ "learning_rate": 3.6984247092188265e-06,
2277
+ "loss": 0.5738,
2278
+ "step": 324
2279
+ },
2280
+ {
2281
+ "epoch": 1.2174320524835989,
2282
+ "grad_norm": 3.0588667392730713,
2283
+ "learning_rate": 3.668535861925509e-06,
2284
+ "loss": 0.9216,
2285
+ "step": 325
2286
+ },
2287
+ {
2288
+ "epoch": 1.2211808809746953,
2289
+ "grad_norm": 2.466240167617798,
2290
+ "learning_rate": 3.6386981371118358e-06,
2291
+ "loss": 0.4853,
2292
+ "step": 326
2293
+ },
2294
+ {
2295
+ "epoch": 1.2249297094657918,
2296
+ "grad_norm": 2.7123985290527344,
2297
+ "learning_rate": 3.6089126804177373e-06,
2298
+ "loss": 0.5507,
2299
+ "step": 327
2300
+ },
2301
+ {
2302
+ "epoch": 1.2286785379568885,
2303
+ "grad_norm": 2.5103797912597656,
2304
+ "learning_rate": 3.5791806354762702e-06,
2305
+ "loss": 0.723,
2306
+ "step": 328
2307
+ },
2308
+ {
2309
+ "epoch": 1.232427366447985,
2310
+ "grad_norm": 2.6206247806549072,
2311
+ "learning_rate": 3.5495031438697103e-06,
2312
+ "loss": 0.6191,
2313
+ "step": 329
2314
+ },
2315
+ {
2316
+ "epoch": 1.2361761949390815,
2317
+ "grad_norm": 2.5603182315826416,
2318
+ "learning_rate": 3.519881345085724e-06,
2319
+ "loss": 0.6315,
2320
+ "step": 330
2321
+ },
2322
+ {
2323
+ "epoch": 1.2399250234301782,
2324
+ "grad_norm": 2.5217936038970947,
2325
+ "learning_rate": 3.4903163764736104e-06,
2326
+ "loss": 0.5989,
2327
+ "step": 331
2328
+ },
2329
+ {
2330
+ "epoch": 1.2436738519212747,
2331
+ "grad_norm": 2.4896228313446045,
2332
+ "learning_rate": 3.4608093732006367e-06,
2333
+ "loss": 0.6737,
2334
+ "step": 332
2335
+ },
2336
+ {
2337
+ "epoch": 1.2474226804123711,
2338
+ "grad_norm": 2.5744566917419434,
2339
+ "learning_rate": 3.4313614682084483e-06,
2340
+ "loss": 0.5537,
2341
+ "step": 333
2342
+ },
2343
+ {
2344
+ "epoch": 1.2511715089034676,
2345
+ "grad_norm": 2.5293214321136475,
2346
+ "learning_rate": 3.401973792169574e-06,
2347
+ "loss": 0.6486,
2348
+ "step": 334
2349
+ },
2350
+ {
2351
+ "epoch": 1.254920337394564,
2352
+ "grad_norm": 2.5666043758392334,
2353
+ "learning_rate": 3.372647473444012e-06,
2354
+ "loss": 0.5285,
2355
+ "step": 335
2356
+ },
2357
+ {
2358
+ "epoch": 1.2586691658856608,
2359
+ "grad_norm": 2.9129676818847656,
2360
+ "learning_rate": 3.343383638035902e-06,
2361
+ "loss": 0.8922,
2362
+ "step": 336
2363
+ },
2364
+ {
2365
+ "epoch": 1.2624179943767573,
2366
+ "grad_norm": 2.596280097961426,
2367
+ "learning_rate": 3.314183409550293e-06,
2368
+ "loss": 0.6352,
2369
+ "step": 337
2370
+ },
2371
+ {
2372
+ "epoch": 1.2661668228678538,
2373
+ "grad_norm": 2.8564982414245605,
2374
+ "learning_rate": 3.285047909150006e-06,
2375
+ "loss": 0.5762,
2376
+ "step": 338
2377
+ },
2378
+ {
2379
+ "epoch": 1.2699156513589505,
2380
+ "grad_norm": 2.6609466075897217,
2381
+ "learning_rate": 3.2559782555125793e-06,
2382
+ "loss": 0.6421,
2383
+ "step": 339
2384
+ },
2385
+ {
2386
+ "epoch": 1.273664479850047,
2387
+ "grad_norm": 2.636430501937866,
2388
+ "learning_rate": 3.226975564787322e-06,
2389
+ "loss": 0.8343,
2390
+ "step": 340
2391
+ },
2392
+ {
2393
+ "epoch": 1.2774133083411434,
2394
+ "grad_norm": 2.764880657196045,
2395
+ "learning_rate": 3.1980409505524546e-06,
2396
+ "loss": 0.8492,
2397
+ "step": 341
2398
+ },
2399
+ {
2400
+ "epoch": 1.28116213683224,
2401
+ "grad_norm": 2.4313178062438965,
2402
+ "learning_rate": 3.1691755237723538e-06,
2403
+ "loss": 0.5935,
2404
+ "step": 342
2405
+ },
2406
+ {
2407
+ "epoch": 1.2849109653233364,
2408
+ "grad_norm": 3.312020778656006,
2409
+ "learning_rate": 3.140380392754901e-06,
2410
+ "loss": 0.7306,
2411
+ "step": 343
2412
+ },
2413
+ {
2414
+ "epoch": 1.2886597938144329,
2415
+ "grad_norm": 2.7336678504943848,
2416
+ "learning_rate": 3.111656663108914e-06,
2417
+ "loss": 0.6683,
2418
+ "step": 344
2419
+ },
2420
+ {
2421
+ "epoch": 1.2924086223055296,
2422
+ "grad_norm": 2.6858274936676025,
2423
+ "learning_rate": 3.083005437701715e-06,
2424
+ "loss": 0.6287,
2425
+ "step": 345
2426
+ },
2427
+ {
2428
+ "epoch": 1.296157450796626,
2429
+ "grad_norm": 2.664910078048706,
2430
+ "learning_rate": 3.054427816616773e-06,
2431
+ "loss": 0.6392,
2432
+ "step": 346
2433
+ },
2434
+ {
2435
+ "epoch": 1.2999062792877225,
2436
+ "grad_norm": 2.3806943893432617,
2437
+ "learning_rate": 3.0259248971114663e-06,
2438
+ "loss": 0.6144,
2439
+ "step": 347
2440
+ },
2441
+ {
2442
+ "epoch": 1.3036551077788192,
2443
+ "grad_norm": 2.755476951599121,
2444
+ "learning_rate": 2.9974977735749596e-06,
2445
+ "loss": 0.6259,
2446
+ "step": 348
2447
+ },
2448
+ {
2449
+ "epoch": 1.3074039362699157,
2450
+ "grad_norm": 2.4375879764556885,
2451
+ "learning_rate": 2.969147537486175e-06,
2452
+ "loss": 0.6063,
2453
+ "step": 349
2454
+ },
2455
+ {
2456
+ "epoch": 1.3111527647610122,
2457
+ "grad_norm": 2.6043853759765625,
2458
+ "learning_rate": 2.9408752773718895e-06,
2459
+ "loss": 0.9196,
2460
+ "step": 350
2461
+ },
2462
+ {
2463
+ "epoch": 1.3149015932521086,
2464
+ "grad_norm": 2.462339401245117,
2465
+ "learning_rate": 2.9126820787649403e-06,
2466
+ "loss": 0.5322,
2467
+ "step": 351
2468
+ },
2469
+ {
2470
+ "epoch": 1.3186504217432051,
2471
+ "grad_norm": 2.3056864738464355,
2472
+ "learning_rate": 2.8845690241625437e-06,
2473
+ "loss": 0.6108,
2474
+ "step": 352
2475
+ },
2476
+ {
2477
+ "epoch": 1.3223992502343018,
2478
+ "grad_norm": 2.394547462463379,
2479
+ "learning_rate": 2.8565371929847286e-06,
2480
+ "loss": 0.6488,
2481
+ "step": 353
2482
+ },
2483
+ {
2484
+ "epoch": 1.3261480787253983,
2485
+ "grad_norm": 3.0996365547180176,
2486
+ "learning_rate": 2.828587661532901e-06,
2487
+ "loss": 0.6514,
2488
+ "step": 354
2489
+ },
2490
+ {
2491
+ "epoch": 1.3298969072164948,
2492
+ "grad_norm": 2.57309889793396,
2493
+ "learning_rate": 2.800721502948506e-06,
2494
+ "loss": 0.614,
2495
+ "step": 355
2496
+ },
2497
+ {
2498
+ "epoch": 1.3336457357075915,
2499
+ "grad_norm": 2.417073965072632,
2500
+ "learning_rate": 2.7729397871718306e-06,
2501
+ "loss": 0.5476,
2502
+ "step": 356
2503
+ },
2504
+ {
2505
+ "epoch": 1.337394564198688,
2506
+ "grad_norm": 2.636715888977051,
2507
+ "learning_rate": 2.7452435809009272e-06,
2508
+ "loss": 0.7593,
2509
+ "step": 357
2510
+ },
2511
+ {
2512
+ "epoch": 1.3411433926897844,
2513
+ "grad_norm": 3.096482992172241,
2514
+ "learning_rate": 2.7176339475506515e-06,
2515
+ "loss": 0.8159,
2516
+ "step": 358
2517
+ },
2518
+ {
2519
+ "epoch": 1.344892221180881,
2520
+ "grad_norm": 2.1638081073760986,
2521
+ "learning_rate": 2.6901119472118253e-06,
2522
+ "loss": 0.5549,
2523
+ "step": 359
2524
+ },
2525
+ {
2526
+ "epoch": 1.3486410496719774,
2527
+ "grad_norm": 2.6135005950927734,
2528
+ "learning_rate": 2.6626786366105493e-06,
2529
+ "loss": 0.6626,
2530
+ "step": 360
2531
+ },
2532
+ {
2533
+ "epoch": 1.352389878163074,
2534
+ "grad_norm": 2.5569889545440674,
2535
+ "learning_rate": 2.635335069067617e-06,
2536
+ "loss": 0.5365,
2537
+ "step": 361
2538
+ },
2539
+ {
2540
+ "epoch": 1.3561387066541706,
2541
+ "grad_norm": 2.9490528106689453,
2542
+ "learning_rate": 2.608082294458074e-06,
2543
+ "loss": 0.9714,
2544
+ "step": 362
2545
+ },
2546
+ {
2547
+ "epoch": 1.359887535145267,
2548
+ "grad_norm": 3.6054556369781494,
2549
+ "learning_rate": 2.5809213591709124e-06,
2550
+ "loss": 0.8305,
2551
+ "step": 363
2552
+ },
2553
+ {
2554
+ "epoch": 1.3636363636363638,
2555
+ "grad_norm": 2.648580312728882,
2556
+ "learning_rate": 2.553853306068888e-06,
2557
+ "loss": 0.7582,
2558
+ "step": 364
2559
+ },
2560
+ {
2561
+ "epoch": 1.3673851921274602,
2562
+ "grad_norm": 2.500655174255371,
2563
+ "learning_rate": 2.5268791744484865e-06,
2564
+ "loss": 0.5625,
2565
+ "step": 365
2566
+ },
2567
+ {
2568
+ "epoch": 1.3711340206185567,
2569
+ "grad_norm": 2.625460147857666,
2570
+ "learning_rate": 2.5000000000000015e-06,
2571
+ "loss": 0.8212,
2572
+ "step": 366
2573
+ },
2574
+ {
2575
+ "epoch": 1.3748828491096532,
2576
+ "grad_norm": 2.444305896759033,
2577
+ "learning_rate": 2.4732168147677927e-06,
2578
+ "loss": 0.582,
2579
+ "step": 367
2580
+ },
2581
+ {
2582
+ "epoch": 1.3786316776007497,
2583
+ "grad_norm": 2.8055059909820557,
2584
+ "learning_rate": 2.446530647110646e-06,
2585
+ "loss": 0.5747,
2586
+ "step": 368
2587
+ },
2588
+ {
2589
+ "epoch": 1.3823805060918464,
2590
+ "grad_norm": 2.7823076248168945,
2591
+ "learning_rate": 2.419942521662285e-06,
2592
+ "loss": 0.5465,
2593
+ "step": 369
2594
+ },
2595
+ {
2596
+ "epoch": 1.3861293345829429,
2597
+ "grad_norm": 3.1207704544067383,
2598
+ "learning_rate": 2.3934534592920416e-06,
2599
+ "loss": 0.8112,
2600
+ "step": 370
2601
+ },
2602
+ {
2603
+ "epoch": 1.3898781630740393,
2604
+ "grad_norm": 2.4543776512145996,
2605
+ "learning_rate": 2.367064477065652e-06,
2606
+ "loss": 0.7124,
2607
+ "step": 371
2608
+ },
2609
+ {
2610
+ "epoch": 1.393626991565136,
2611
+ "grad_norm": 2.6401026248931885,
2612
+ "learning_rate": 2.3407765882062024e-06,
2613
+ "loss": 0.6856,
2614
+ "step": 372
2615
+ },
2616
+ {
2617
+ "epoch": 1.3973758200562325,
2618
+ "grad_norm": 2.729567527770996,
2619
+ "learning_rate": 2.314590802055232e-06,
2620
+ "loss": 0.6516,
2621
+ "step": 373
2622
+ },
2623
+ {
2624
+ "epoch": 1.401124648547329,
2625
+ "grad_norm": 2.283984899520874,
2626
+ "learning_rate": 2.2885081240339813e-06,
2627
+ "loss": 0.4082,
2628
+ "step": 374
2629
+ },
2630
+ {
2631
+ "epoch": 1.4048734770384255,
2632
+ "grad_norm": 2.7015268802642822,
2633
+ "learning_rate": 2.262529555604774e-06,
2634
+ "loss": 0.7567,
2635
+ "step": 375
2636
+ },
2637
+ {
2638
+ "epoch": 1.408622305529522,
2639
+ "grad_norm": 2.3578450679779053,
2640
+ "learning_rate": 2.2366560942325833e-06,
2641
+ "loss": 0.6722,
2642
+ "step": 376
2643
+ },
2644
+ {
2645
+ "epoch": 1.4123711340206184,
2646
+ "grad_norm": 2.6497104167938232,
2647
+ "learning_rate": 2.2108887333467172e-06,
2648
+ "loss": 0.6284,
2649
+ "step": 377
2650
+ },
2651
+ {
2652
+ "epoch": 1.4161199625117151,
2653
+ "grad_norm": 2.5546791553497314,
2654
+ "learning_rate": 2.1852284623026906e-06,
2655
+ "loss": 0.5851,
2656
+ "step": 378
2657
+ },
2658
+ {
2659
+ "epoch": 1.4198687910028116,
2660
+ "grad_norm": 2.486978530883789,
2661
+ "learning_rate": 2.159676266344222e-06,
2662
+ "loss": 0.6988,
2663
+ "step": 379
2664
+ },
2665
+ {
2666
+ "epoch": 1.423617619493908,
2667
+ "grad_norm": 2.8536384105682373,
2668
+ "learning_rate": 2.1342331265654194e-06,
2669
+ "loss": 0.7193,
2670
+ "step": 380
2671
+ },
2672
+ {
2673
+ "epoch": 1.4273664479850048,
2674
+ "grad_norm": 2.9641923904418945,
2675
+ "learning_rate": 2.108900019873103e-06,
2676
+ "loss": 0.6805,
2677
+ "step": 381
2678
+ },
2679
+ {
2680
+ "epoch": 1.4311152764761013,
2681
+ "grad_norm": 2.463860034942627,
2682
+ "learning_rate": 2.0836779189492925e-06,
2683
+ "loss": 0.7775,
2684
+ "step": 382
2685
+ },
2686
+ {
2687
+ "epoch": 1.4348641049671977,
2688
+ "grad_norm": 2.7579569816589355,
2689
+ "learning_rate": 2.0585677922138696e-06,
2690
+ "loss": 0.6252,
2691
+ "step": 383
2692
+ },
2693
+ {
2694
+ "epoch": 1.4386129334582942,
2695
+ "grad_norm": 2.5609865188598633,
2696
+ "learning_rate": 2.033570603787391e-06,
2697
+ "loss": 0.6355,
2698
+ "step": 384
2699
+ },
2700
+ {
2701
+ "epoch": 1.4423617619493907,
2702
+ "grad_norm": 2.9254584312438965,
2703
+ "learning_rate": 2.0086873134540626e-06,
2704
+ "loss": 0.6698,
2705
+ "step": 385
2706
+ },
2707
+ {
2708
+ "epoch": 1.4461105904404874,
2709
+ "grad_norm": 3.046320915222168,
2710
+ "learning_rate": 1.9839188766249024e-06,
2711
+ "loss": 0.6857,
2712
+ "step": 386
2713
+ },
2714
+ {
2715
+ "epoch": 1.4498594189315839,
2716
+ "grad_norm": 2.7485857009887695,
2717
+ "learning_rate": 1.959266244301047e-06,
2718
+ "loss": 0.8151,
2719
+ "step": 387
2720
+ },
2721
+ {
2722
+ "epoch": 1.4536082474226804,
2723
+ "grad_norm": 2.6440517902374268,
2724
+ "learning_rate": 1.9347303630372373e-06,
2725
+ "loss": 0.63,
2726
+ "step": 388
2727
+ },
2728
+ {
2729
+ "epoch": 1.457357075913777,
2730
+ "grad_norm": 2.842188596725464,
2731
+ "learning_rate": 1.910312174905477e-06,
2732
+ "loss": 0.5113,
2733
+ "step": 389
2734
+ },
2735
+ {
2736
+ "epoch": 1.4611059044048735,
2737
+ "grad_norm": 2.662004232406616,
2738
+ "learning_rate": 1.886012617458864e-06,
2739
+ "loss": 0.7212,
2740
+ "step": 390
2741
+ },
2742
+ {
2743
+ "epoch": 1.46485473289597,
2744
+ "grad_norm": 3.0616824626922607,
2745
+ "learning_rate": 1.8618326236955908e-06,
2746
+ "loss": 0.4773,
2747
+ "step": 391
2748
+ },
2749
+ {
2750
+ "epoch": 1.4686035613870665,
2751
+ "grad_norm": 2.796537399291992,
2752
+ "learning_rate": 1.8377731220231144e-06,
2753
+ "loss": 0.7543,
2754
+ "step": 392
2755
+ },
2756
+ {
2757
+ "epoch": 1.472352389878163,
2758
+ "grad_norm": 2.2921605110168457,
2759
+ "learning_rate": 1.8138350362225193e-06,
2760
+ "loss": 0.5149,
2761
+ "step": 393
2762
+ },
2763
+ {
2764
+ "epoch": 1.4761012183692597,
2765
+ "grad_norm": 3.190627098083496,
2766
+ "learning_rate": 1.7900192854130465e-06,
2767
+ "loss": 0.7386,
2768
+ "step": 394
2769
+ },
2770
+ {
2771
+ "epoch": 1.4798500468603561,
2772
+ "grad_norm": 2.5983688831329346,
2773
+ "learning_rate": 1.7663267840167936e-06,
2774
+ "loss": 0.6061,
2775
+ "step": 395
2776
+ },
2777
+ {
2778
+ "epoch": 1.4835988753514526,
2779
+ "grad_norm": 2.47511887550354,
2780
+ "learning_rate": 1.7427584417236194e-06,
2781
+ "loss": 0.4628,
2782
+ "step": 396
2783
+ },
2784
+ {
2785
+ "epoch": 1.4873477038425493,
2786
+ "grad_norm": 2.6179940700531006,
2787
+ "learning_rate": 1.7193151634562071e-06,
2788
+ "loss": 0.7502,
2789
+ "step": 397
2790
+ },
2791
+ {
2792
+ "epoch": 1.4910965323336458,
2793
+ "grad_norm": 2.2079148292541504,
2794
+ "learning_rate": 1.695997849335319e-06,
2795
+ "loss": 0.7195,
2796
+ "step": 398
2797
+ },
2798
+ {
2799
+ "epoch": 1.4948453608247423,
2800
+ "grad_norm": 2.3705711364746094,
2801
+ "learning_rate": 1.672807394645236e-06,
2802
+ "loss": 0.679,
2803
+ "step": 399
2804
+ },
2805
+ {
2806
+ "epoch": 1.4985941893158388,
2807
+ "grad_norm": 2.5409562587738037,
2808
+ "learning_rate": 1.6497446897993885e-06,
2809
+ "loss": 0.6188,
2810
+ "step": 400
2811
+ },
2812
+ {
2813
+ "epoch": 1.5023430178069352,
2814
+ "grad_norm": 2.718851089477539,
2815
+ "learning_rate": 1.6268106203061628e-06,
2816
+ "loss": 0.4311,
2817
+ "step": 401
2818
+ },
2819
+ {
2820
+ "epoch": 1.5060918462980317,
2821
+ "grad_norm": 2.4857470989227295,
2822
+ "learning_rate": 1.6040060667348995e-06,
2823
+ "loss": 0.5452,
2824
+ "step": 402
2825
+ },
2826
+ {
2827
+ "epoch": 1.5098406747891284,
2828
+ "grad_norm": 2.272148609161377,
2829
+ "learning_rate": 1.581331904682089e-06,
2830
+ "loss": 0.6673,
2831
+ "step": 403
2832
+ },
2833
+ {
2834
+ "epoch": 1.513589503280225,
2835
+ "grad_norm": 2.118818759918213,
2836
+ "learning_rate": 1.5587890047377512e-06,
2837
+ "loss": 0.474,
2838
+ "step": 404
2839
+ },
2840
+ {
2841
+ "epoch": 1.5173383317713216,
2842
+ "grad_norm": 2.6302151679992676,
2843
+ "learning_rate": 1.5363782324520033e-06,
2844
+ "loss": 0.7036,
2845
+ "step": 405
2846
+ },
2847
+ {
2848
+ "epoch": 1.521087160262418,
2849
+ "grad_norm": 2.78167724609375,
2850
+ "learning_rate": 1.5141004483018323e-06,
2851
+ "loss": 0.5538,
2852
+ "step": 406
2853
+ },
2854
+ {
2855
+ "epoch": 1.5248359887535146,
2856
+ "grad_norm": 2.7051169872283936,
2857
+ "learning_rate": 1.4919565076580577e-06,
2858
+ "loss": 0.6842,
2859
+ "step": 407
2860
+ },
2861
+ {
2862
+ "epoch": 1.528584817244611,
2863
+ "grad_norm": 2.7234108448028564,
2864
+ "learning_rate": 1.4699472607524785e-06,
2865
+ "loss": 0.6473,
2866
+ "step": 408
2867
+ },
2868
+ {
2869
+ "epoch": 1.5323336457357075,
2870
+ "grad_norm": 2.808751106262207,
2871
+ "learning_rate": 1.4480735526452427e-06,
2872
+ "loss": 0.7719,
2873
+ "step": 409
2874
+ },
2875
+ {
2876
+ "epoch": 1.536082474226804,
2877
+ "grad_norm": 2.5814692974090576,
2878
+ "learning_rate": 1.426336223192386e-06,
2879
+ "loss": 0.6786,
2880
+ "step": 410
2881
+ },
2882
+ {
2883
+ "epoch": 1.5398313027179007,
2884
+ "grad_norm": 2.6848583221435547,
2885
+ "learning_rate": 1.4047361070135996e-06,
2886
+ "loss": 0.7245,
2887
+ "step": 411
2888
+ },
2889
+ {
2890
+ "epoch": 1.5435801312089972,
2891
+ "grad_norm": 2.525697708129883,
2892
+ "learning_rate": 1.3832740334601692e-06,
2893
+ "loss": 0.5269,
2894
+ "step": 412
2895
+ },
2896
+ {
2897
+ "epoch": 1.5473289597000939,
2898
+ "grad_norm": 2.4603147506713867,
2899
+ "learning_rate": 1.3619508265831445e-06,
2900
+ "loss": 0.7593,
2901
+ "step": 413
2902
+ },
2903
+ {
2904
+ "epoch": 1.5510777881911904,
2905
+ "grad_norm": 2.712268114089966,
2906
+ "learning_rate": 1.340767305101694e-06,
2907
+ "loss": 0.8538,
2908
+ "step": 414
2909
+ },
2910
+ {
2911
+ "epoch": 1.5548266166822868,
2912
+ "grad_norm": 2.391826629638672,
2913
+ "learning_rate": 1.319724282371664e-06,
2914
+ "loss": 0.7118,
2915
+ "step": 415
2916
+ },
2917
+ {
2918
+ "epoch": 1.5585754451733833,
2919
+ "grad_norm": 2.6861860752105713,
2920
+ "learning_rate": 1.2988225663543601e-06,
2921
+ "loss": 0.6899,
2922
+ "step": 416
2923
+ },
2924
+ {
2925
+ "epoch": 1.5623242736644798,
2926
+ "grad_norm": 3.0465526580810547,
2927
+ "learning_rate": 1.2780629595855203e-06,
2928
+ "loss": 0.6717,
2929
+ "step": 417
2930
+ },
2931
+ {
2932
+ "epoch": 1.5660731021555763,
2933
+ "grad_norm": 2.586912155151367,
2934
+ "learning_rate": 1.257446259144494e-06,
2935
+ "loss": 0.6787,
2936
+ "step": 418
2937
+ },
2938
+ {
2939
+ "epoch": 1.569821930646673,
2940
+ "grad_norm": 2.4131391048431396,
2941
+ "learning_rate": 1.2369732566236508e-06,
2942
+ "loss": 0.7233,
2943
+ "step": 419
2944
+ },
2945
+ {
2946
+ "epoch": 1.5735707591377694,
2947
+ "grad_norm": 2.3727686405181885,
2948
+ "learning_rate": 1.2166447380979801e-06,
2949
+ "loss": 0.7571,
2950
+ "step": 420
2951
+ },
2952
+ {
2953
+ "epoch": 1.577319587628866,
2954
+ "grad_norm": 2.27282452583313,
2955
+ "learning_rate": 1.1964614840949002e-06,
2956
+ "loss": 0.5189,
2957
+ "step": 421
2958
+ },
2959
+ {
2960
+ "epoch": 1.5810684161199626,
2961
+ "grad_norm": 2.7784337997436523,
2962
+ "learning_rate": 1.1764242695643075e-06,
2963
+ "loss": 0.4099,
2964
+ "step": 422
2965
+ },
2966
+ {
2967
+ "epoch": 1.584817244611059,
2968
+ "grad_norm": 2.5282742977142334,
2969
+ "learning_rate": 1.1565338638488117e-06,
2970
+ "loss": 0.5868,
2971
+ "step": 423
2972
+ },
2973
+ {
2974
+ "epoch": 1.5885660731021556,
2975
+ "grad_norm": 2.49729061126709,
2976
+ "learning_rate": 1.1367910306541918e-06,
2977
+ "loss": 0.6738,
2978
+ "step": 424
2979
+ },
2980
+ {
2981
+ "epoch": 1.592314901593252,
2982
+ "grad_norm": 2.5107173919677734,
2983
+ "learning_rate": 1.1171965280200831e-06,
2984
+ "loss": 0.5341,
2985
+ "step": 425
2986
+ },
2987
+ {
2988
+ "epoch": 1.5960637300843485,
2989
+ "grad_norm": 2.601574659347534,
2990
+ "learning_rate": 1.097751108290867e-06,
2991
+ "loss": 0.692,
2992
+ "step": 426
2993
+ },
2994
+ {
2995
+ "epoch": 1.599812558575445,
2996
+ "grad_norm": 2.5476865768432617,
2997
+ "learning_rate": 1.078455518086784e-06,
2998
+ "loss": 0.7559,
2999
+ "step": 427
3000
+ },
3001
+ {
3002
+ "epoch": 1.6035613870665417,
3003
+ "grad_norm": 2.2659873962402344,
3004
+ "learning_rate": 1.0593104982752645e-06,
3005
+ "loss": 0.5147,
3006
+ "step": 428
3007
+ },
3008
+ {
3009
+ "epoch": 1.6073102155576382,
3010
+ "grad_norm": 2.6162917613983154,
3011
+ "learning_rate": 1.0403167839424883e-06,
3012
+ "loss": 0.7862,
3013
+ "step": 429
3014
+ },
3015
+ {
3016
+ "epoch": 1.611059044048735,
3017
+ "grad_norm": 3.0038373470306396,
3018
+ "learning_rate": 1.0214751043651582e-06,
3019
+ "loss": 0.7098,
3020
+ "step": 430
3021
+ },
3022
+ {
3023
+ "epoch": 1.6148078725398314,
3024
+ "grad_norm": 2.810269832611084,
3025
+ "learning_rate": 1.0027861829824953e-06,
3026
+ "loss": 0.4501,
3027
+ "step": 431
3028
+ },
3029
+ {
3030
+ "epoch": 1.6185567010309279,
3031
+ "grad_norm": 2.2872512340545654,
3032
+ "learning_rate": 9.842507373684646e-07,
3033
+ "loss": 0.6829,
3034
+ "step": 432
3035
+ },
3036
+ {
3037
+ "epoch": 1.6223055295220243,
3038
+ "grad_norm": 2.7606618404388428,
3039
+ "learning_rate": 9.658694792042284e-07,
3040
+ "loss": 0.8641,
3041
+ "step": 433
3042
+ },
3043
+ {
3044
+ "epoch": 1.6260543580131208,
3045
+ "grad_norm": 2.647319793701172,
3046
+ "learning_rate": 9.476431142508097e-07,
3047
+ "loss": 0.684,
3048
+ "step": 434
3049
+ },
3050
+ {
3051
+ "epoch": 1.6298031865042173,
3052
+ "grad_norm": 2.3936805725097656,
3053
+ "learning_rate": 9.295723423220049e-07,
3054
+ "loss": 0.4689,
3055
+ "step": 435
3056
+ },
3057
+ {
3058
+ "epoch": 1.633552014995314,
3059
+ "grad_norm": 2.8151843547821045,
3060
+ "learning_rate": 9.116578572575091e-07,
3061
+ "loss": 0.7088,
3062
+ "step": 436
3063
+ },
3064
+ {
3065
+ "epoch": 1.6373008434864105,
3066
+ "grad_norm": 2.6258625984191895,
3067
+ "learning_rate": 8.939003468962726e-07,
3068
+ "loss": 0.7019,
3069
+ "step": 437
3070
+ },
3071
+ {
3072
+ "epoch": 1.6410496719775072,
3073
+ "grad_norm": 2.601217031478882,
3074
+ "learning_rate": 8.763004930500979e-07,
3075
+ "loss": 0.5881,
3076
+ "step": 438
3077
+ },
3078
+ {
3079
+ "epoch": 1.6447985004686037,
3080
+ "grad_norm": 2.325805187225342,
3081
+ "learning_rate": 8.58858971477457e-07,
3082
+ "loss": 0.6775,
3083
+ "step": 439
3084
+ },
3085
+ {
3086
+ "epoch": 1.6485473289597001,
3087
+ "grad_norm": 2.8295209407806396,
3088
+ "learning_rate": 8.415764518575415e-07,
3089
+ "loss": 0.7012,
3090
+ "step": 440
3091
+ },
3092
+ {
3093
+ "epoch": 1.6522961574507966,
3094
+ "grad_norm": 2.421982765197754,
3095
+ "learning_rate": 8.244535977645584e-07,
3096
+ "loss": 0.6516,
3097
+ "step": 441
3098
+ },
3099
+ {
3100
+ "epoch": 1.656044985941893,
3101
+ "grad_norm": 2.5072197914123535,
3102
+ "learning_rate": 8.074910666422475e-07,
3103
+ "loss": 0.7382,
3104
+ "step": 442
3105
+ },
3106
+ {
3107
+ "epoch": 1.6597938144329896,
3108
+ "grad_norm": 2.841768264770508,
3109
+ "learning_rate": 7.906895097786338e-07,
3110
+ "loss": 0.5838,
3111
+ "step": 443
3112
+ },
3113
+ {
3114
+ "epoch": 1.6635426429240863,
3115
+ "grad_norm": 2.738243818283081,
3116
+ "learning_rate": 7.740495722810271e-07,
3117
+ "loss": 0.7085,
3118
+ "step": 444
3119
+ },
3120
+ {
3121
+ "epoch": 1.6672914714151827,
3122
+ "grad_norm": 2.763880491256714,
3123
+ "learning_rate": 7.575718930512516e-07,
3124
+ "loss": 0.6332,
3125
+ "step": 445
3126
+ },
3127
+ {
3128
+ "epoch": 1.6710402999062794,
3129
+ "grad_norm": 2.3609251976013184,
3130
+ "learning_rate": 7.412571047611156e-07,
3131
+ "loss": 0.5682,
3132
+ "step": 446
3133
+ },
3134
+ {
3135
+ "epoch": 1.674789128397376,
3136
+ "grad_norm": 3.1101462841033936,
3137
+ "learning_rate": 7.25105833828113e-07,
3138
+ "loss": 0.8626,
3139
+ "step": 447
3140
+ },
3141
+ {
3142
+ "epoch": 1.6785379568884724,
3143
+ "grad_norm": 2.56254506111145,
3144
+ "learning_rate": 7.091187003913802e-07,
3145
+ "loss": 0.733,
3146
+ "step": 448
3147
+ },
3148
+ {
3149
+ "epoch": 1.6822867853795689,
3150
+ "grad_norm": 2.901610851287842,
3151
+ "learning_rate": 6.932963182878821e-07,
3152
+ "loss": 0.7838,
3153
+ "step": 449
3154
+ },
3155
+ {
3156
+ "epoch": 1.6860356138706654,
3157
+ "grad_norm": 2.393528699874878,
3158
+ "learning_rate": 6.776392950288397e-07,
3159
+ "loss": 0.6754,
3160
+ "step": 450
3161
+ },
3162
+ {
3163
+ "epoch": 1.6897844423617618,
3164
+ "grad_norm": 2.7382218837738037,
3165
+ "learning_rate": 6.621482317764105e-07,
3166
+ "loss": 0.7144,
3167
+ "step": 451
3168
+ },
3169
+ {
3170
+ "epoch": 1.6935332708528583,
3171
+ "grad_norm": 2.618114471435547,
3172
+ "learning_rate": 6.468237233206043e-07,
3173
+ "loss": 0.7003,
3174
+ "step": 452
3175
+ },
3176
+ {
3177
+ "epoch": 1.697282099343955,
3178
+ "grad_norm": 2.418599843978882,
3179
+ "learning_rate": 6.316663580564425e-07,
3180
+ "loss": 0.6957,
3181
+ "step": 453
3182
+ },
3183
+ {
3184
+ "epoch": 1.7010309278350515,
3185
+ "grad_norm": 2.5395448207855225,
3186
+ "learning_rate": 6.166767179613691e-07,
3187
+ "loss": 0.5987,
3188
+ "step": 454
3189
+ },
3190
+ {
3191
+ "epoch": 1.7047797563261482,
3192
+ "grad_norm": 2.5443437099456787,
3193
+ "learning_rate": 6.018553785729075e-07,
3194
+ "loss": 0.4977,
3195
+ "step": 455
3196
+ },
3197
+ {
3198
+ "epoch": 1.7085285848172447,
3199
+ "grad_norm": 2.4432942867279053,
3200
+ "learning_rate": 5.872029089665588e-07,
3201
+ "loss": 0.675,
3202
+ "step": 456
3203
+ },
3204
+ {
3205
+ "epoch": 1.7122774133083412,
3206
+ "grad_norm": 2.7112233638763428,
3207
+ "learning_rate": 5.727198717339511e-07,
3208
+ "loss": 0.7934,
3209
+ "step": 457
3210
+ },
3211
+ {
3212
+ "epoch": 1.7160262417994376,
3213
+ "grad_norm": 2.9048352241516113,
3214
+ "learning_rate": 5.584068229612422e-07,
3215
+ "loss": 0.7687,
3216
+ "step": 458
3217
+ },
3218
+ {
3219
+ "epoch": 1.7197750702905341,
3220
+ "grad_norm": 2.5611398220062256,
3221
+ "learning_rate": 5.442643122077673e-07,
3222
+ "loss": 0.516,
3223
+ "step": 459
3224
+ },
3225
+ {
3226
+ "epoch": 1.7235238987816306,
3227
+ "grad_norm": 2.6895134449005127,
3228
+ "learning_rate": 5.302928824849335e-07,
3229
+ "loss": 0.662,
3230
+ "step": 460
3231
+ },
3232
+ {
3233
+ "epoch": 1.7272727272727273,
3234
+ "grad_norm": 2.621039390563965,
3235
+ "learning_rate": 5.164930702353782e-07,
3236
+ "loss": 0.6618,
3237
+ "step": 461
3238
+ },
3239
+ {
3240
+ "epoch": 1.7310215557638238,
3241
+ "grad_norm": 2.5842714309692383,
3242
+ "learning_rate": 5.028654053123666e-07,
3243
+ "loss": 0.5826,
3244
+ "step": 462
3245
+ },
3246
+ {
3247
+ "epoch": 1.7347703842549205,
3248
+ "grad_norm": 2.4873173236846924,
3249
+ "learning_rate": 4.894104109594466e-07,
3250
+ "loss": 0.6776,
3251
+ "step": 463
3252
+ },
3253
+ {
3254
+ "epoch": 1.738519212746017,
3255
+ "grad_norm": 2.2931952476501465,
3256
+ "learning_rate": 4.7612860379036674e-07,
3257
+ "loss": 0.6094,
3258
+ "step": 464
3259
+ },
3260
+ {
3261
+ "epoch": 1.7422680412371134,
3262
+ "grad_norm": 2.699720859527588,
3263
+ "learning_rate": 4.6302049376922843e-07,
3264
+ "loss": 0.5347,
3265
+ "step": 465
3266
+ },
3267
+ {
3268
+ "epoch": 1.74601686972821,
3269
+ "grad_norm": 3.027738571166992,
3270
+ "learning_rate": 4.500865841909169e-07,
3271
+ "loss": 0.6739,
3272
+ "step": 466
3273
+ },
3274
+ {
3275
+ "epoch": 1.7497656982193064,
3276
+ "grad_norm": 2.582227945327759,
3277
+ "learning_rate": 4.373273716617682e-07,
3278
+ "loss": 0.547,
3279
+ "step": 467
3280
+ },
3281
+ {
3282
+ "epoch": 1.7535145267104029,
3283
+ "grad_norm": 3.154305934906006,
3284
+ "learning_rate": 4.247433460805067e-07,
3285
+ "loss": 0.7603,
3286
+ "step": 468
3287
+ },
3288
+ {
3289
+ "epoch": 1.7572633552014996,
3290
+ "grad_norm": 2.7173681259155273,
3291
+ "learning_rate": 4.123349906194357e-07,
3292
+ "loss": 0.6546,
3293
+ "step": 469
3294
+ },
3295
+ {
3296
+ "epoch": 1.761012183692596,
3297
+ "grad_norm": 2.6075384616851807,
3298
+ "learning_rate": 4.001027817058789e-07,
3299
+ "loss": 0.6589,
3300
+ "step": 470
3301
+ },
3302
+ {
3303
+ "epoch": 1.7647610121836927,
3304
+ "grad_norm": 2.7403595447540283,
3305
+ "learning_rate": 3.8804718900389673e-07,
3306
+ "loss": 0.7657,
3307
+ "step": 471
3308
+ },
3309
+ {
3310
+ "epoch": 1.7685098406747892,
3311
+ "grad_norm": 2.6404995918273926,
3312
+ "learning_rate": 3.7616867539624733e-07,
3313
+ "loss": 0.4362,
3314
+ "step": 472
3315
+ },
3316
+ {
3317
+ "epoch": 1.7722586691658857,
3318
+ "grad_norm": 2.6141321659088135,
3319
+ "learning_rate": 3.6446769696661445e-07,
3320
+ "loss": 0.5542,
3321
+ "step": 473
3322
+ },
3323
+ {
3324
+ "epoch": 1.7760074976569822,
3325
+ "grad_norm": 2.7053098678588867,
3326
+ "learning_rate": 3.5294470298209817e-07,
3327
+ "loss": 0.7215,
3328
+ "step": 474
3329
+ },
3330
+ {
3331
+ "epoch": 1.7797563261480787,
3332
+ "grad_norm": 2.3503482341766357,
3333
+ "learning_rate": 3.416001358759635e-07,
3334
+ "loss": 0.5368,
3335
+ "step": 475
3336
+ },
3337
+ {
3338
+ "epoch": 1.7835051546391751,
3339
+ "grad_norm": 2.7156739234924316,
3340
+ "learning_rate": 3.304344312306529e-07,
3341
+ "loss": 0.6804,
3342
+ "step": 476
3343
+ },
3344
+ {
3345
+ "epoch": 1.7872539831302718,
3346
+ "grad_norm": 2.5638582706451416,
3347
+ "learning_rate": 3.194480177610604e-07,
3348
+ "loss": 0.8309,
3349
+ "step": 477
3350
+ },
3351
+ {
3352
+ "epoch": 1.7910028116213683,
3353
+ "grad_norm": 2.8452188968658447,
3354
+ "learning_rate": 3.08641317298074e-07,
3355
+ "loss": 0.5907,
3356
+ "step": 478
3357
+ },
3358
+ {
3359
+ "epoch": 1.794751640112465,
3360
+ "grad_norm": 2.5471601486206055,
3361
+ "learning_rate": 2.980147447723775e-07,
3362
+ "loss": 0.6524,
3363
+ "step": 479
3364
+ },
3365
+ {
3366
+ "epoch": 1.7985004686035615,
3367
+ "grad_norm": 2.443708658218384,
3368
+ "learning_rate": 2.8756870819851736e-07,
3369
+ "loss": 0.5538,
3370
+ "step": 480
3371
+ },
3372
+ {
3373
+ "epoch": 1.802249297094658,
3374
+ "grad_norm": 2.7727556228637695,
3375
+ "learning_rate": 2.7730360865923954e-07,
3376
+ "loss": 0.6459,
3377
+ "step": 481
3378
+ },
3379
+ {
3380
+ "epoch": 1.8059981255857545,
3381
+ "grad_norm": 3.0080816745758057,
3382
+ "learning_rate": 2.672198402900883e-07,
3383
+ "loss": 0.7243,
3384
+ "step": 482
3385
+ },
3386
+ {
3387
+ "epoch": 1.809746954076851,
3388
+ "grad_norm": 2.2730140686035156,
3389
+ "learning_rate": 2.573177902642726e-07,
3390
+ "loss": 0.7811,
3391
+ "step": 483
3392
+ },
3393
+ {
3394
+ "epoch": 1.8134957825679474,
3395
+ "grad_norm": 2.5842466354370117,
3396
+ "learning_rate": 2.475978387778e-07,
3397
+ "loss": 0.595,
3398
+ "step": 484
3399
+ },
3400
+ {
3401
+ "epoch": 1.817244611059044,
3402
+ "grad_norm": 2.4688827991485596,
3403
+ "learning_rate": 2.380603590348829e-07,
3404
+ "loss": 0.6941,
3405
+ "step": 485
3406
+ },
3407
+ {
3408
+ "epoch": 1.8209934395501406,
3409
+ "grad_norm": 2.346224546432495,
3410
+ "learning_rate": 2.2870571723360212e-07,
3411
+ "loss": 0.7242,
3412
+ "step": 486
3413
+ },
3414
+ {
3415
+ "epoch": 1.824742268041237,
3416
+ "grad_norm": 2.721781015396118,
3417
+ "learning_rate": 2.1953427255185122e-07,
3418
+ "loss": 0.7004,
3419
+ "step": 487
3420
+ },
3421
+ {
3422
+ "epoch": 1.8284910965323338,
3423
+ "grad_norm": 2.539914846420288,
3424
+ "learning_rate": 2.1054637713354586e-07,
3425
+ "loss": 0.6182,
3426
+ "step": 488
3427
+ },
3428
+ {
3429
+ "epoch": 1.8322399250234302,
3430
+ "grad_norm": 2.6623215675354004,
3431
+ "learning_rate": 2.0174237607510138e-07,
3432
+ "loss": 0.4989,
3433
+ "step": 489
3434
+ },
3435
+ {
3436
+ "epoch": 1.8359887535145267,
3437
+ "grad_norm": 2.411862373352051,
3438
+ "learning_rate": 1.9312260741218114e-07,
3439
+ "loss": 0.4213,
3440
+ "step": 490
3441
+ },
3442
+ {
3443
+ "epoch": 1.8397375820056232,
3444
+ "grad_norm": 2.1537928581237793,
3445
+ "learning_rate": 1.8468740210672077e-07,
3446
+ "loss": 0.5353,
3447
+ "step": 491
3448
+ },
3449
+ {
3450
+ "epoch": 1.8434864104967197,
3451
+ "grad_norm": 3.083552598953247,
3452
+ "learning_rate": 1.7643708403422055e-07,
3453
+ "loss": 1.1005,
3454
+ "step": 492
3455
+ },
3456
+ {
3457
+ "epoch": 1.8472352389878162,
3458
+ "grad_norm": 2.445988178253174,
3459
+ "learning_rate": 1.6837196997130434e-07,
3460
+ "loss": 0.6158,
3461
+ "step": 493
3462
+ },
3463
+ {
3464
+ "epoch": 1.8509840674789129,
3465
+ "grad_norm": 2.336162567138672,
3466
+ "learning_rate": 1.6049236958356475e-07,
3467
+ "loss": 0.649,
3468
+ "step": 494
3469
+ },
3470
+ {
3471
+ "epoch": 1.8547328959700093,
3472
+ "grad_norm": 2.544160842895508,
3473
+ "learning_rate": 1.5279858541366876e-07,
3474
+ "loss": 0.7222,
3475
+ "step": 495
3476
+ },
3477
+ {
3478
+ "epoch": 1.858481724461106,
3479
+ "grad_norm": 2.5376548767089844,
3480
+ "learning_rate": 1.4529091286973994e-07,
3481
+ "loss": 0.7403,
3482
+ "step": 496
3483
+ },
3484
+ {
3485
+ "epoch": 1.8622305529522025,
3486
+ "grad_norm": 3.008063554763794,
3487
+ "learning_rate": 1.3796964021402072e-07,
3488
+ "loss": 0.7811,
3489
+ "step": 497
3490
+ },
3491
+ {
3492
+ "epoch": 1.865979381443299,
3493
+ "grad_norm": 2.640939712524414,
3494
+ "learning_rate": 1.3083504855180007e-07,
3495
+ "loss": 0.6864,
3496
+ "step": 498
3497
+ },
3498
+ {
3499
+ "epoch": 1.8697282099343955,
3500
+ "grad_norm": 2.6732051372528076,
3501
+ "learning_rate": 1.2388741182062348e-07,
3502
+ "loss": 0.6969,
3503
+ "step": 499
3504
+ },
3505
+ {
3506
+ "epoch": 1.873477038425492,
3507
+ "grad_norm": 2.688314199447632,
3508
+ "learning_rate": 1.1712699677977224e-07,
3509
+ "loss": 0.659,
3510
+ "step": 500
3511
+ },
3512
+ {
3513
+ "epoch": 1.8772258669165884,
3514
+ "grad_norm": 2.5495784282684326,
3515
+ "learning_rate": 1.1055406300002347e-07,
3516
+ "loss": 0.5468,
3517
+ "step": 501
3518
+ },
3519
+ {
3520
+ "epoch": 1.8809746954076851,
3521
+ "grad_norm": 2.6933047771453857,
3522
+ "learning_rate": 1.0416886285368188e-07,
3523
+ "loss": 0.5204,
3524
+ "step": 502
3525
+ },
3526
+ {
3527
+ "epoch": 1.8847235238987816,
3528
+ "grad_norm": 2.7378575801849365,
3529
+ "learning_rate": 9.797164150489035e-08,
3530
+ "loss": 0.5859,
3531
+ "step": 503
3532
+ },
3533
+ {
3534
+ "epoch": 1.8884723523898783,
3535
+ "grad_norm": 2.0356202125549316,
3536
+ "learning_rate": 9.1962636900218e-08,
3537
+ "loss": 0.3406,
3538
+ "step": 504
3539
+ },
3540
+ {
3541
+ "epoch": 1.8922211808809748,
3542
+ "grad_norm": 3.0371510982513428,
3543
+ "learning_rate": 8.614207975952083e-08,
3544
+ "loss": 0.7767,
3545
+ "step": 505
3546
+ },
3547
+ {
3548
+ "epoch": 1.8959700093720713,
3549
+ "grad_norm": 2.813620090484619,
3550
+ "learning_rate": 8.0510193567086e-08,
3551
+ "loss": 0.5616,
3552
+ "step": 506
3553
+ },
3554
+ {
3555
+ "epoch": 1.8997188378631678,
3556
+ "grad_norm": 2.557776689529419,
3557
+ "learning_rate": 7.5067194563051e-08,
3558
+ "loss": 0.5433,
3559
+ "step": 507
3560
+ },
3561
+ {
3562
+ "epoch": 1.9034676663542642,
3563
+ "grad_norm": 2.6855971813201904,
3564
+ "learning_rate": 6.981329173509909e-08,
3565
+ "loss": 0.6334,
3566
+ "step": 508
3567
+ },
3568
+ {
3569
+ "epoch": 1.9072164948453607,
3570
+ "grad_norm": 2.539196729660034,
3571
+ "learning_rate": 6.474868681043578e-08,
3572
+ "loss": 0.6286,
3573
+ "step": 509
3574
+ },
3575
+ {
3576
+ "epoch": 1.9109653233364574,
3577
+ "grad_norm": 2.1722402572631836,
3578
+ "learning_rate": 5.987357424804441e-08,
3579
+ "loss": 0.5581,
3580
+ "step": 510
3581
+ },
3582
+ {
3583
+ "epoch": 1.914714151827554,
3584
+ "grad_norm": 2.4900224208831787,
3585
+ "learning_rate": 5.518814123121885e-08,
3586
+ "loss": 0.6132,
3587
+ "step": 511
3588
+ },
3589
+ {
3590
+ "epoch": 1.9184629803186504,
3591
+ "grad_norm": 2.708784818649292,
3592
+ "learning_rate": 5.0692567660375334e-08,
3593
+ "loss": 0.6154,
3594
+ "step": 512
3595
+ },
3596
+ {
3597
+ "epoch": 1.922211808809747,
3598
+ "grad_norm": 2.573993444442749,
3599
+ "learning_rate": 4.638702614614854e-08,
3600
+ "loss": 0.6727,
3601
+ "step": 513
3602
+ },
3603
+ {
3604
+ "epoch": 1.9259606373008435,
3605
+ "grad_norm": 2.826658248901367,
3606
+ "learning_rate": 4.227168200276077e-08,
3607
+ "loss": 0.7038,
3608
+ "step": 514
3609
+ },
3610
+ {
3611
+ "epoch": 1.92970946579194,
3612
+ "grad_norm": 2.657594680786133,
3613
+ "learning_rate": 3.834669324167428e-08,
3614
+ "loss": 0.7022,
3615
+ "step": 515
3616
+ },
3617
+ {
3618
+ "epoch": 1.9334582942830365,
3619
+ "grad_norm": 2.5230019092559814,
3620
+ "learning_rate": 3.4612210565528323e-08,
3621
+ "loss": 0.5854,
3622
+ "step": 516
3623
+ },
3624
+ {
3625
+ "epoch": 1.937207122774133,
3626
+ "grad_norm": 2.966693162918091,
3627
+ "learning_rate": 3.10683773623488e-08,
3628
+ "loss": 0.6926,
3629
+ "step": 517
3630
+ },
3631
+ {
3632
+ "epoch": 1.9409559512652295,
3633
+ "grad_norm": 2.689201593399048,
3634
+ "learning_rate": 2.7715329700044315e-08,
3635
+ "loss": 0.7031,
3636
+ "step": 518
3637
+ },
3638
+ {
3639
+ "epoch": 1.9447047797563262,
3640
+ "grad_norm": 2.46077036857605,
3641
+ "learning_rate": 2.455319632118147e-08,
3642
+ "loss": 0.5426,
3643
+ "step": 519
3644
+ },
3645
+ {
3646
+ "epoch": 1.9484536082474226,
3647
+ "grad_norm": 2.4496285915374756,
3648
+ "learning_rate": 2.158209863804217e-08,
3649
+ "loss": 0.5547,
3650
+ "step": 520
3651
+ },
3652
+ {
3653
+ "epoch": 1.9522024367385193,
3654
+ "grad_norm": 2.5419907569885254,
3655
+ "learning_rate": 1.8802150727962876e-08,
3656
+ "loss": 0.5622,
3657
+ "step": 521
3658
+ },
3659
+ {
3660
+ "epoch": 1.9559512652296158,
3661
+ "grad_norm": 2.629879951477051,
3662
+ "learning_rate": 1.6213459328950355e-08,
3663
+ "loss": 0.6756,
3664
+ "step": 522
3665
+ },
3666
+ {
3667
+ "epoch": 1.9597000937207123,
3668
+ "grad_norm": 2.748396396636963,
3669
+ "learning_rate": 1.3816123835588835e-08,
3670
+ "loss": 0.5403,
3671
+ "step": 523
3672
+ },
3673
+ {
3674
+ "epoch": 1.9634489222118088,
3675
+ "grad_norm": 2.49360728263855,
3676
+ "learning_rate": 1.161023629522029e-08,
3677
+ "loss": 0.735,
3678
+ "step": 524
3679
+ },
3680
+ {
3681
+ "epoch": 1.9671977507029053,
3682
+ "grad_norm": 2.472134828567505,
3683
+ "learning_rate": 9.595881404411145e-09,
3684
+ "loss": 0.5665,
3685
+ "step": 525
3686
+ },
3687
+ {
3688
+ "epoch": 1.9709465791940017,
3689
+ "grad_norm": 2.7294187545776367,
3690
+ "learning_rate": 7.773136505700995e-09,
3691
+ "loss": 0.8548,
3692
+ "step": 526
3693
+ },
3694
+ {
3695
+ "epoch": 1.9746954076850984,
3696
+ "grad_norm": 2.79746675491333,
3697
+ "learning_rate": 6.142071584630538e-09,
3698
+ "loss": 0.7157,
3699
+ "step": 527
3700
+ },
3701
+ {
3702
+ "epoch": 1.978444236176195,
3703
+ "grad_norm": 2.4330995082855225,
3704
+ "learning_rate": 4.702749267057604e-09,
3705
+ "loss": 0.5992,
3706
+ "step": 528
3707
+ },
3708
+ {
3709
+ "epoch": 1.9821930646672916,
3710
+ "grad_norm": 2.6152493953704834,
3711
+ "learning_rate": 3.4552248167507576e-09,
3712
+ "loss": 0.6798,
3713
+ "step": 529
3714
+ },
3715
+ {
3716
+ "epoch": 1.985941893158388,
3717
+ "grad_norm": 2.38468861579895,
3718
+ "learning_rate": 2.3995461332676496e-09,
3719
+ "loss": 0.5985,
3720
+ "step": 530
3721
+ },
3722
+ {
3723
+ "epoch": 1.9896907216494846,
3724
+ "grad_norm": 2.6697652339935303,
3725
+ "learning_rate": 1.5357537501159425e-09,
3726
+ "loss": 0.5737,
3727
+ "step": 531
3728
+ },
3729
+ {
3730
+ "epoch": 1.993439550140581,
3731
+ "grad_norm": 2.0764174461364746,
3732
+ "learning_rate": 8.638808331973281e-10,
3733
+ "loss": 0.5335,
3734
+ "step": 532
3735
+ },
3736
+ {
3737
+ "epoch": 1.9971883786316775,
3738
+ "grad_norm": 2.559661626815796,
3739
+ "learning_rate": 3.8395317953354717e-10,
3740
+ "loss": 0.6069,
3741
+ "step": 533
3742
+ },
3743
+ {
3744
+ "epoch": 2.0,
3745
+ "grad_norm": 3.018411636352539,
3746
+ "learning_rate": 9.598921627607116e-11,
3747
+ "loss": 0.5042,
3748
+ "step": 534
3749
+ }
3750
+ ],
3751
+ "logging_steps": 1,
3752
+ "max_steps": 534,
3753
+ "num_input_tokens_seen": 0,
3754
+ "num_train_epochs": 2,
3755
+ "save_steps": 500,
3756
+ "stateful_callbacks": {
3757
+ "TrainerControl": {
3758
+ "args": {
3759
+ "should_epoch_stop": false,
3760
+ "should_evaluate": false,
3761
+ "should_log": false,
3762
+ "should_save": true,
3763
+ "should_training_stop": true
3764
+ },
3765
+ "attributes": {}
3766
+ }
3767
+ },
3768
+ "total_flos": 3.243263714603827e+16,
3769
+ "train_batch_size": 6,
3770
+ "trial_name": null,
3771
+ "trial_params": null
3772
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dea00308b4a5bdeafd42e0f635440f49101360a33138babcf903b6306ca6a8ca
3
+ size 5496
vocab.json ADDED
The diff for this file is too large to render. See raw diff