Hugging Face's logo

psp-dada
/
Qwen2.5-VL-7B-Instruct-SENTINEL

Image-Text-to-Text
Transformers
Safetensors
English
lora
conversational
Model card Files
xet
Community
1
psp_dada commited on
Commit
aa2784f
·
1 Parent(s): 0b45db8

add file

Browse files
Files changed (16) hide show
  1. .gitattributes +1 -0
  2. adapter_config.json +120 -0
  3. adapter_model.safetensors +3 -0
  4. added_tokens.json +24 -0
  5. all_results.json +8 -0
  6. chat_template.json +3 -0
  7. merges.txt +0 -0
  8. preprocessor_config.json +29 -0
  9. special_tokens_map.json +31 -0
  10. tokenizer.json +3 -0
  11. tokenizer_config.json +210 -0
  12. train_results.json +8 -0
  13. trainer_log.jsonl +110 -0
  14. trainer_state.json +1678 -0
  15. training_args.bin +3 -0
  16. vocab.json +0 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
adapter_config.json ADDED
@@ -0,0 +1,120 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "Qwen/Qwen2.5-VL-7B-Instruct",
5
+ "bias": "none",
6
+ "corda_config": null,
7
+ "eva_config": null,
8
+ "exclude_modules": null,
9
+ "fan_in_fan_out": false,
10
+ "inference_mode": true,
11
+ "init_lora_weights": true,
12
+ "layer_replication": null,
13
+ "layers_pattern": null,
14
+ "layers_to_transform": null,
15
+ "loftq_config": {},
16
+ "lora_alpha": 256,
17
+ "lora_bias": false,
18
+ "lora_dropout": 0.0,
19
+ "megatron_config": null,
20
+ "megatron_core": "megatron.core",
21
+ "modules_to_save": null,
22
+ "peft_type": "LORA",
23
+ "r": 128,
24
+ "rank_pattern": {},
25
+ "revision": null,
26
+ "target_modules": [
27
+ "layers.10.mlp.gate_proj",
28
+ "layers.8.mlp.up_proj",
29
+ "layers.10.mlp.down_proj",
30
+ "layers.15.mlp.gate_proj",
31
+ "layers.17.mlp.down_proj",
32
+ "layers.14.mlp.gate_proj",
33
+ "layers.22.mlp.down_proj",
34
+ "layers.21.mlp.down_proj",
35
+ "layers.27.mlp.down_proj",
36
+ "layers.23.mlp.down_proj",
37
+ "layers.26.mlp.down_proj",
38
+ "layers.9.mlp.up_proj",
39
+ "layers.15.mlp.down_proj",
40
+ "layers.11.mlp.down_proj",
41
+ "k_proj",
42
+ "layers.7.mlp.up_proj",
43
+ "layers.16.mlp.gate_proj",
44
+ "layers.21.mlp.gate_proj",
45
+ "layers.14.mlp.down_proj",
46
+ "layers.17.mlp.gate_proj",
47
+ "layers.8.mlp.gate_proj",
48
+ "layers.2.mlp.up_proj",
49
+ "layers.27.mlp.up_proj",
50
+ "layers.6.mlp.gate_proj",
51
+ "layers.5.mlp.up_proj",
52
+ "layers.16.mlp.up_proj",
53
+ "layers.19.mlp.down_proj",
54
+ "layers.11.mlp.gate_proj",
55
+ "layers.22.mlp.gate_proj",
56
+ "layers.3.mlp.up_proj",
57
+ "layers.24.mlp.down_proj",
58
+ "layers.23.mlp.gate_proj",
59
+ "layers.4.mlp.gate_proj",
60
+ "layers.12.mlp.gate_proj",
61
+ "layers.0.mlp.down_proj",
62
+ "layers.26.mlp.up_proj",
63
+ "layers.18.mlp.gate_proj",
64
+ "layers.0.mlp.gate_proj",
65
+ "layers.13.mlp.gate_proj",
66
+ "layers.2.mlp.down_proj",
67
+ "layers.24.mlp.gate_proj",
68
+ "layers.26.mlp.gate_proj",
69
+ "layers.10.mlp.up_proj",
70
+ "layers.27.mlp.gate_proj",
71
+ "layers.20.mlp.down_proj",
72
+ "layers.4.mlp.down_proj",
73
+ "layers.3.mlp.down_proj",
74
+ "layers.7.mlp.down_proj",
75
+ "layers.11.mlp.up_proj",
76
+ "layers.9.mlp.down_proj",
77
+ "layers.1.mlp.gate_proj",
78
+ "layers.22.mlp.up_proj",
79
+ "layers.4.mlp.up_proj",
80
+ "layers.15.mlp.up_proj",
81
+ "layers.6.mlp.up_proj",
82
+ "o_proj",
83
+ "layers.18.mlp.up_proj",
84
+ "layers.23.mlp.up_proj",
85
+ "layers.5.mlp.gate_proj",
86
+ "layers.19.mlp.up_proj",
87
+ "layers.12.mlp.up_proj",
88
+ "layers.19.mlp.gate_proj",
89
+ "layers.16.mlp.down_proj",
90
+ "layers.21.mlp.up_proj",
91
+ "layers.18.mlp.down_proj",
92
+ "q_proj",
93
+ "layers.0.mlp.up_proj",
94
+ "layers.13.mlp.down_proj",
95
+ "layers.12.mlp.down_proj",
96
+ "layers.6.mlp.down_proj",
97
+ "layers.25.mlp.up_proj",
98
+ "layers.20.mlp.gate_proj",
99
+ "layers.2.mlp.gate_proj",
100
+ "layers.13.mlp.up_proj",
101
+ "layers.3.mlp.gate_proj",
102
+ "layers.25.mlp.gate_proj",
103
+ "layers.1.mlp.up_proj",
104
+ "layers.5.mlp.down_proj",
105
+ "layers.7.mlp.gate_proj",
106
+ "layers.25.mlp.down_proj",
107
+ "layers.14.mlp.up_proj",
108
+ "layers.17.mlp.up_proj",
109
+ "layers.20.mlp.up_proj",
110
+ "layers.1.mlp.down_proj",
111
+ "layers.9.mlp.gate_proj",
112
+ "v_proj",
113
+ "layers.8.mlp.down_proj",
114
+ "layers.24.mlp.up_proj"
115
+ ],
116
+ "task_type": "CAUSAL_LM",
117
+ "trainable_token_indices": null,
118
+ "use_dora": false,
119
+ "use_rslora": false
120
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:01a78a23f8a11a51d8f50be1bce4218f5d3a17d4319fb701739b3d845ca960a6
3
+ size 645976488
added_tokens.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</tool_call>": 151658,
3
+ "<tool_call>": 151657,
4
+ "<|box_end|>": 151649,
5
+ "<|box_start|>": 151648,
6
+ "<|endoftext|>": 151643,
7
+ "<|file_sep|>": 151664,
8
+ "<|fim_middle|>": 151660,
9
+ "<|fim_pad|>": 151662,
10
+ "<|fim_prefix|>": 151659,
11
+ "<|fim_suffix|>": 151661,
12
+ "<|im_end|>": 151645,
13
+ "<|im_start|>": 151644,
14
+ "<|image_pad|>": 151655,
15
+ "<|object_ref_end|>": 151647,
16
+ "<|object_ref_start|>": 151646,
17
+ "<|quad_end|>": 151651,
18
+ "<|quad_start|>": 151650,
19
+ "<|repo_name|>": 151663,
20
+ "<|video_pad|>": 151656,
21
+ "<|vision_end|>": 151653,
22
+ "<|vision_pad|>": 151654,
23
+ "<|vision_start|>": 151652
24
+ }
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.9965714285714286,
3
+ "total_flos": 4.2490128934476186e+17,
4
+ "train_loss": 0.666442971710765,
5
+ "train_runtime": 4023.4671,
6
+ "train_samples_per_second": 1.74,
7
+ "train_steps_per_second": 0.027
8
+ }
chat_template.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
3
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
preprocessor_config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_convert_rgb": true,
3
+ "do_normalize": true,
4
+ "do_rescale": true,
5
+ "do_resize": true,
6
+ "image_mean": [
7
+ 0.48145466,
8
+ 0.4578275,
9
+ 0.40821073
10
+ ],
11
+ "image_processor_type": "Qwen2VLImageProcessor",
12
+ "image_std": [
13
+ 0.26862954,
14
+ 0.26130258,
15
+ 0.27577711
16
+ ],
17
+ "max_pixels": 12845056,
18
+ "merge_size": 2,
19
+ "min_pixels": 3136,
20
+ "patch_size": 14,
21
+ "processor_class": "Qwen2_5_VLProcessor",
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "longest_edge": 12845056,
26
+ "shortest_edge": 3136
27
+ },
28
+ "temporal_patch_size": 2
29
+ }
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": "<|im_end|>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "pad_token": {
25
+ "content": "<|endoftext|>",
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 sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
3
+ size 11421896
tokenizer_config.json ADDED
@@ -0,0 +1,210 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ },
182
+ "additional_special_tokens": [
183
+ "<|im_start|>",
184
+ "<|im_end|>",
185
+ "<|object_ref_start|>",
186
+ "<|object_ref_end|>",
187
+ "<|box_start|>",
188
+ "<|box_end|>",
189
+ "<|quad_start|>",
190
+ "<|quad_end|>",
191
+ "<|vision_start|>",
192
+ "<|vision_end|>",
193
+ "<|vision_pad|>",
194
+ "<|image_pad|>",
195
+ "<|video_pad|>"
196
+ ],
197
+ "bos_token": null,
198
+ "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}",
199
+ "clean_up_tokenization_spaces": false,
200
+ "eos_token": "<|im_end|>",
201
+ "errors": "replace",
202
+ "extra_special_tokens": {},
203
+ "model_max_length": 131072,
204
+ "pad_token": "<|endoftext|>",
205
+ "padding_side": "right",
206
+ "processor_class": "Qwen2_5_VLProcessor",
207
+ "split_special_tokens": false,
208
+ "tokenizer_class": "Qwen2Tokenizer",
209
+ "unk_token": null
210
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.9965714285714286,
3
+ "total_flos": 4.2490128934476186e+17,
4
+ "train_loss": 0.666442971710765,
5
+ "train_runtime": 4023.4671,
6
+ "train_samples_per_second": 1.74,
7
+ "train_steps_per_second": 0.027
8
+ }
trainer_log.jsonl ADDED
@@ -0,0 +1,110 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"current_steps": 1, "total_steps": 109, "loss": 0.6931, "accuracy": 0.0, "lr": 3e-06, "epoch": 0.009142857142857144, "percentage": 0.92, "elapsed_time": "0:00:30", "remaining_time": "0:55:31"}
2
+ {"current_steps": 2, "total_steps": 109, "loss": 0.6929, "accuracy": 0.40625, "lr": 2.9993770144857767e-06, "epoch": 0.018285714285714287, "percentage": 1.83, "elapsed_time": "0:01:04", "remaining_time": "0:57:23"}
3
+ {"current_steps": 3, "total_steps": 109, "loss": 0.6925, "accuracy": 0.390625, "lr": 2.997508575424375e-06, "epoch": 0.027428571428571427, "percentage": 2.75, "elapsed_time": "0:01:36", "remaining_time": "0:57:01"}
4
+ {"current_steps": 4, "total_steps": 109, "loss": 0.6941, "accuracy": 0.421875, "lr": 2.9943962348297537e-06, "epoch": 0.036571428571428574, "percentage": 3.67, "elapsed_time": "0:02:08", "remaining_time": "0:56:12"}
5
+ {"current_steps": 5, "total_steps": 109, "loss": 0.6915, "accuracy": 0.5, "lr": 2.9900425779593876e-06, "epoch": 0.045714285714285714, "percentage": 4.59, "elapsed_time": "0:02:42", "remaining_time": "0:56:28"}
6
+ {"current_steps": 6, "total_steps": 109, "loss": 0.6925, "accuracy": 0.421875, "lr": 2.9844512211668286e-06, "epoch": 0.054857142857142854, "percentage": 5.5, "elapsed_time": "0:03:19", "remaining_time": "0:57:00"}
7
+ {"current_steps": 7, "total_steps": 109, "loss": 0.6891, "accuracy": 0.53125, "lr": 2.977626808897792e-06, "epoch": 0.064, "percentage": 6.42, "elapsed_time": "0:03:53", "remaining_time": "0:56:46"}
8
+ {"current_steps": 8, "total_steps": 109, "loss": 0.6949, "accuracy": 0.359375, "lr": 2.9695750098322613e-06, "epoch": 0.07314285714285715, "percentage": 7.34, "elapsed_time": "0:04:29", "remaining_time": "0:56:46"}
9
+ {"current_steps": 9, "total_steps": 109, "loss": 0.6907, "accuracy": 0.46875, "lr": 2.9603025121758102e-06, "epoch": 0.08228571428571428, "percentage": 8.26, "elapsed_time": "0:05:06", "remaining_time": "0:56:48"}
10
+ {"current_steps": 10, "total_steps": 109, "loss": 0.6899, "accuracy": 0.53125, "lr": 2.9498170181040663e-06, "epoch": 0.09142857142857143, "percentage": 9.17, "elapsed_time": "0:05:46", "remaining_time": "0:57:06"}
11
+ {"current_steps": 11, "total_steps": 109, "loss": 0.692, "accuracy": 0.484375, "lr": 2.938127237364918e-06, "epoch": 0.10057142857142858, "percentage": 10.09, "elapsed_time": "0:06:21", "remaining_time": "0:56:38"}
12
+ {"current_steps": 12, "total_steps": 109, "loss": 0.6886, "accuracy": 0.59375, "lr": 2.925242880043786e-06, "epoch": 0.10971428571428571, "percentage": 11.01, "elapsed_time": "0:06:54", "remaining_time": "0:55:49"}
13
+ {"current_steps": 13, "total_steps": 109, "loss": 0.6912, "accuracy": 0.46875, "lr": 2.911174648497964e-06, "epoch": 0.11885714285714286, "percentage": 11.93, "elapsed_time": "0:07:28", "remaining_time": "0:55:13"}
14
+ {"current_steps": 14, "total_steps": 109, "loss": 0.6924, "accuracy": 0.46875, "lr": 2.895934228466738e-06, "epoch": 0.128, "percentage": 12.84, "elapsed_time": "0:08:07", "remaining_time": "0:55:09"}
15
+ {"current_steps": 15, "total_steps": 109, "loss": 0.692, "accuracy": 0.515625, "lr": 2.879534279364654e-06, "epoch": 0.13714285714285715, "percentage": 13.76, "elapsed_time": "0:08:47", "remaining_time": "0:55:08"}
16
+ {"current_steps": 16, "total_steps": 109, "loss": 0.6953, "accuracy": 0.390625, "lr": 2.8619884237660125e-06, "epoch": 0.1462857142857143, "percentage": 14.68, "elapsed_time": "0:09:26", "remaining_time": "0:54:55"}
17
+ {"current_steps": 17, "total_steps": 109, "loss": 0.6885, "accuracy": 0.46875, "lr": 2.843311236089309e-06, "epoch": 0.15542857142857142, "percentage": 15.6, "elapsed_time": "0:10:04", "remaining_time": "0:54:29"}
18
+ {"current_steps": 18, "total_steps": 109, "loss": 0.6869, "accuracy": 0.546875, "lr": 2.8235182304910364e-06, "epoch": 0.16457142857142856, "percentage": 16.51, "elapsed_time": "0:10:37", "remaining_time": "0:53:44"}
19
+ {"current_steps": 19, "total_steps": 109, "loss": 0.6882, "accuracy": 0.53125, "lr": 2.8026258479788888e-06, "epoch": 0.1737142857142857, "percentage": 17.43, "elapsed_time": "0:11:17", "remaining_time": "0:53:30"}
20
+ {"current_steps": 20, "total_steps": 109, "loss": 0.6856, "accuracy": 0.59375, "lr": 2.780651442755083e-06, "epoch": 0.18285714285714286, "percentage": 18.35, "elapsed_time": "0:11:55", "remaining_time": "0:53:03"}
21
+ {"current_steps": 21, "total_steps": 109, "loss": 0.6852, "accuracy": 0.578125, "lr": 2.7576132678011365e-06, "epoch": 0.192, "percentage": 19.27, "elapsed_time": "0:12:31", "remaining_time": "0:52:30"}
22
+ {"current_steps": 22, "total_steps": 109, "loss": 0.6849, "accuracy": 0.484375, "lr": 2.7335304597160764e-06, "epoch": 0.20114285714285715, "percentage": 20.18, "elapsed_time": "0:13:11", "remaining_time": "0:52:10"}
23
+ {"current_steps": 23, "total_steps": 109, "loss": 0.6857, "accuracy": 0.515625, "lr": 2.7084230228206746e-06, "epoch": 0.2102857142857143, "percentage": 21.1, "elapsed_time": "0:13:46", "remaining_time": "0:51:31"}
24
+ {"current_steps": 24, "total_steps": 109, "loss": 0.6854, "accuracy": 0.484375, "lr": 2.6823118125409112e-06, "epoch": 0.21942857142857142, "percentage": 22.02, "elapsed_time": "0:14:28", "remaining_time": "0:51:17"}
25
+ {"current_steps": 25, "total_steps": 109, "loss": 0.6858, "accuracy": 0.5625, "lr": 2.6552185180844704e-06, "epoch": 0.22857142857142856, "percentage": 22.94, "elapsed_time": "0:15:09", "remaining_time": "0:50:56"}
26
+ {"current_steps": 26, "total_steps": 109, "loss": 0.6822, "accuracy": 0.59375, "lr": 2.6271656444246578e-06, "epoch": 0.2377142857142857, "percentage": 23.85, "elapsed_time": "0:15:38", "remaining_time": "0:49:56"}
27
+ {"current_steps": 27, "total_steps": 109, "loss": 0.6912, "accuracy": 0.484375, "lr": 2.598176493606703e-06, "epoch": 0.24685714285714286, "percentage": 24.77, "elapsed_time": "0:16:13", "remaining_time": "0:49:16"}
28
+ {"current_steps": 28, "total_steps": 109, "loss": 0.6845, "accuracy": 0.625, "lr": 2.568275145391978e-06, "epoch": 0.256, "percentage": 25.69, "elapsed_time": "0:16:48", "remaining_time": "0:48:36"}
29
+ {"current_steps": 29, "total_steps": 109, "loss": 0.6784, "accuracy": 0.625, "lr": 2.5374864372562077e-06, "epoch": 0.2651428571428571, "percentage": 26.61, "elapsed_time": "0:17:29", "remaining_time": "0:48:13"}
30
+ {"current_steps": 30, "total_steps": 109, "loss": 0.6809, "accuracy": 0.625, "lr": 2.505835943758286e-06, "epoch": 0.2742857142857143, "percentage": 27.52, "elapsed_time": "0:18:00", "remaining_time": "0:47:25"}
31
+ {"current_steps": 31, "total_steps": 109, "loss": 0.678, "accuracy": 0.53125, "lr": 2.4733499552968357e-06, "epoch": 0.2834285714285714, "percentage": 28.44, "elapsed_time": "0:18:39", "remaining_time": "0:46:55"}
32
+ {"current_steps": 32, "total_steps": 109, "loss": 0.669, "accuracy": 0.71875, "lr": 2.440055456272159e-06, "epoch": 0.2925714285714286, "percentage": 29.36, "elapsed_time": "0:19:12", "remaining_time": "0:46:12"}
33
+ {"current_steps": 33, "total_steps": 109, "loss": 0.678, "accuracy": 0.640625, "lr": 2.4059801026717166e-06, "epoch": 0.3017142857142857, "percentage": 30.28, "elapsed_time": "0:19:50", "remaining_time": "0:45:41"}
34
+ {"current_steps": 34, "total_steps": 109, "loss": 0.6784, "accuracy": 0.5625, "lr": 2.3711521990977554e-06, "epoch": 0.31085714285714283, "percentage": 31.19, "elapsed_time": "0:20:23", "remaining_time": "0:44:58"}
35
+ {"current_steps": 35, "total_steps": 109, "loss": 0.6812, "accuracy": 0.53125, "lr": 2.3356006752561658e-06, "epoch": 0.32, "percentage": 32.11, "elapsed_time": "0:20:55", "remaining_time": "0:44:14"}
36
+ {"current_steps": 36, "total_steps": 109, "loss": 0.6691, "accuracy": 0.65625, "lr": 2.299355061926096e-06, "epoch": 0.3291428571428571, "percentage": 33.03, "elapsed_time": "0:21:27", "remaining_time": "0:43:31"}
37
+ {"current_steps": 37, "total_steps": 109, "loss": 0.6801, "accuracy": 0.65625, "lr": 2.262445466430292e-06, "epoch": 0.3382857142857143, "percentage": 33.94, "elapsed_time": "0:21:58", "remaining_time": "0:42:45"}
38
+ {"current_steps": 38, "total_steps": 109, "loss": 0.6679, "accuracy": 0.609375, "lr": 2.2249025476265262e-06, "epoch": 0.3474285714285714, "percentage": 34.86, "elapsed_time": "0:22:30", "remaining_time": "0:42:03"}
39
+ {"current_steps": 39, "total_steps": 109, "loss": 0.6718, "accuracy": 0.703125, "lr": 2.1867574904409007e-06, "epoch": 0.3565714285714286, "percentage": 35.78, "elapsed_time": "0:23:01", "remaining_time": "0:41:20"}
40
+ {"current_steps": 40, "total_steps": 109, "loss": 0.6774, "accuracy": 0.578125, "lr": 2.1480419799641695e-06, "epoch": 0.3657142857142857, "percentage": 36.7, "elapsed_time": "0:23:35", "remaining_time": "0:40:41"}
41
+ {"current_steps": 41, "total_steps": 109, "loss": 0.6628, "accuracy": 0.671875, "lr": 2.1087881751326035e-06, "epoch": 0.37485714285714283, "percentage": 37.61, "elapsed_time": "0:24:10", "remaining_time": "0:40:06"}
42
+ {"current_steps": 42, "total_steps": 109, "loss": 0.6767, "accuracy": 0.59375, "lr": 2.0690286820152535e-06, "epoch": 0.384, "percentage": 38.53, "elapsed_time": "0:24:42", "remaining_time": "0:39:25"}
43
+ {"current_steps": 43, "total_steps": 109, "loss": 0.6676, "accuracy": 0.578125, "lr": 2.028796526729806e-06, "epoch": 0.3931428571428571, "percentage": 39.45, "elapsed_time": "0:25:17", "remaining_time": "0:38:49"}
44
+ {"current_steps": 44, "total_steps": 109, "loss": 0.6733, "accuracy": 0.609375, "lr": 1.9881251280095263e-06, "epoch": 0.4022857142857143, "percentage": 40.37, "elapsed_time": "0:25:54", "remaining_time": "0:38:15"}
45
+ {"current_steps": 45, "total_steps": 109, "loss": 0.6639, "accuracy": 0.703125, "lr": 1.9470482694440755e-06, "epoch": 0.4114285714285714, "percentage": 41.28, "elapsed_time": "0:26:32", "remaining_time": "0:37:45"}
46
+ {"current_steps": 46, "total_steps": 109, "loss": 0.6574, "accuracy": 0.6875, "lr": 1.9056000714172617e-06, "epoch": 0.4205714285714286, "percentage": 42.2, "elapsed_time": "0:27:07", "remaining_time": "0:37:08"}
47
+ {"current_steps": 47, "total_steps": 109, "loss": 0.6709, "accuracy": 0.53125, "lr": 1.8638149627650335e-06, "epoch": 0.4297142857142857, "percentage": 43.12, "elapsed_time": "0:27:37", "remaining_time": "0:36:26"}
48
+ {"current_steps": 48, "total_steps": 109, "loss": 0.6599, "accuracy": 0.59375, "lr": 1.8217276521772582e-06, "epoch": 0.43885714285714283, "percentage": 44.04, "elapsed_time": "0:28:15", "remaining_time": "0:35:54"}
49
+ {"current_steps": 49, "total_steps": 109, "loss": 0.6695, "accuracy": 0.5625, "lr": 1.7793730993670408e-06, "epoch": 0.448, "percentage": 44.95, "elapsed_time": "0:28:44", "remaining_time": "0:35:12"}
50
+ {"current_steps": 50, "total_steps": 109, "loss": 0.6651, "accuracy": 0.65625, "lr": 1.736786486031531e-06, "epoch": 0.45714285714285713, "percentage": 45.87, "elapsed_time": "0:29:19", "remaining_time": "0:34:35"}
51
+ {"current_steps": 51, "total_steps": 109, "loss": 0.6756, "accuracy": 0.578125, "lr": 1.6940031866283395e-06, "epoch": 0.4662857142857143, "percentage": 46.79, "elapsed_time": "0:29:52", "remaining_time": "0:33:58"}
52
+ {"current_steps": 52, "total_steps": 109, "loss": 0.6582, "accuracy": 0.609375, "lr": 1.6510587389918377e-06, "epoch": 0.4754285714285714, "percentage": 47.71, "elapsed_time": "0:30:28", "remaining_time": "0:33:24"}
53
+ {"current_steps": 53, "total_steps": 109, "loss": 0.6594, "accuracy": 0.671875, "lr": 1.6079888148137507e-06, "epoch": 0.4845714285714286, "percentage": 48.62, "elapsed_time": "0:31:02", "remaining_time": "0:32:47"}
54
+ {"current_steps": 54, "total_steps": 109, "loss": 0.6668, "accuracy": 0.640625, "lr": 1.564829190012561e-06, "epoch": 0.4937142857142857, "percentage": 49.54, "elapsed_time": "0:31:35", "remaining_time": "0:32:10"}
55
+ {"current_steps": 55, "total_steps": 109, "loss": 0.6678, "accuracy": 0.578125, "lr": 1.521615715016336e-06, "epoch": 0.5028571428571429, "percentage": 50.46, "elapsed_time": "0:32:09", "remaining_time": "0:31:34"}
56
+ {"current_steps": 56, "total_steps": 109, "loss": 0.6781, "accuracy": 0.546875, "lr": 1.4783842849836645e-06, "epoch": 0.512, "percentage": 51.38, "elapsed_time": "0:32:45", "remaining_time": "0:31:00"}
57
+ {"current_steps": 57, "total_steps": 109, "loss": 0.6551, "accuracy": 0.59375, "lr": 1.435170809987439e-06, "epoch": 0.5211428571428571, "percentage": 52.29, "elapsed_time": "0:33:16", "remaining_time": "0:30:21"}
58
+ {"current_steps": 58, "total_steps": 109, "loss": 0.6596, "accuracy": 0.703125, "lr": 1.3920111851862494e-06, "epoch": 0.5302857142857142, "percentage": 53.21, "elapsed_time": "0:33:57", "remaining_time": "0:29:51"}
59
+ {"current_steps": 59, "total_steps": 109, "loss": 0.6581, "accuracy": 0.6875, "lr": 1.3489412610081626e-06, "epoch": 0.5394285714285715, "percentage": 54.13, "elapsed_time": "0:34:32", "remaining_time": "0:29:16"}
60
+ {"current_steps": 60, "total_steps": 109, "loss": 0.6533, "accuracy": 0.671875, "lr": 1.3059968133716607e-06, "epoch": 0.5485714285714286, "percentage": 55.05, "elapsed_time": "0:35:08", "remaining_time": "0:28:41"}
61
+ {"current_steps": 61, "total_steps": 109, "loss": 0.6609, "accuracy": 0.640625, "lr": 1.2632135139684691e-06, "epoch": 0.5577142857142857, "percentage": 55.96, "elapsed_time": "0:35:41", "remaining_time": "0:28:04"}
62
+ {"current_steps": 62, "total_steps": 109, "loss": 0.6653, "accuracy": 0.515625, "lr": 1.2206269006329595e-06, "epoch": 0.5668571428571428, "percentage": 56.88, "elapsed_time": "0:36:18", "remaining_time": "0:27:31"}
63
+ {"current_steps": 63, "total_steps": 109, "loss": 0.6769, "accuracy": 0.53125, "lr": 1.178272347822742e-06, "epoch": 0.576, "percentage": 57.8, "elapsed_time": "0:36:56", "remaining_time": "0:26:58"}
64
+ {"current_steps": 64, "total_steps": 109, "loss": 0.6602, "accuracy": 0.65625, "lr": 1.1361850372349668e-06, "epoch": 0.5851428571428572, "percentage": 58.72, "elapsed_time": "0:37:29", "remaining_time": "0:26:21"}
65
+ {"current_steps": 65, "total_steps": 109, "loss": 0.6563, "accuracy": 0.609375, "lr": 1.0943999285827381e-06, "epoch": 0.5942857142857143, "percentage": 59.63, "elapsed_time": "0:38:05", "remaining_time": "0:25:46"}
66
+ {"current_steps": 66, "total_steps": 109, "loss": 0.6761, "accuracy": 0.484375, "lr": 1.0529517305559246e-06, "epoch": 0.6034285714285714, "percentage": 60.55, "elapsed_time": "0:38:39", "remaining_time": "0:25:10"}
67
+ {"current_steps": 67, "total_steps": 109, "loss": 0.6571, "accuracy": 0.59375, "lr": 1.0118748719904738e-06, "epoch": 0.6125714285714285, "percentage": 61.47, "elapsed_time": "0:39:16", "remaining_time": "0:24:36"}
68
+ {"current_steps": 68, "total_steps": 109, "loss": 0.6603, "accuracy": 0.640625, "lr": 9.712034732701943e-07, "epoch": 0.6217142857142857, "percentage": 62.39, "elapsed_time": "0:40:01", "remaining_time": "0:24:08"}
69
+ {"current_steps": 69, "total_steps": 109, "loss": 0.6431, "accuracy": 0.65625, "lr": 9.309713179847465e-07, "epoch": 0.6308571428571429, "percentage": 63.3, "elapsed_time": "0:40:46", "remaining_time": "0:23:38"}
70
+ {"current_steps": 70, "total_steps": 109, "loss": 0.6577, "accuracy": 0.609375, "lr": 8.912118248673967e-07, "epoch": 0.64, "percentage": 64.22, "elapsed_time": "0:41:30", "remaining_time": "0:23:07"}
71
+ {"current_steps": 71, "total_steps": 109, "loss": 0.6637, "accuracy": 0.640625, "lr": 8.519580200358309e-07, "epoch": 0.6491428571428571, "percentage": 65.14, "elapsed_time": "0:42:11", "remaining_time": "0:22:34"}
72
+ {"current_steps": 72, "total_steps": 109, "loss": 0.634, "accuracy": 0.640625, "lr": 8.132425095591e-07, "epoch": 0.6582857142857143, "percentage": 66.06, "elapsed_time": "0:42:49", "remaining_time": "0:22:00"}
73
+ {"current_steps": 73, "total_steps": 109, "loss": 0.644, "accuracy": 0.65625, "lr": 7.750974523734742e-07, "epoch": 0.6674285714285715, "percentage": 66.97, "elapsed_time": "0:43:24", "remaining_time": "0:21:24"}
74
+ {"current_steps": 74, "total_steps": 109, "loss": 0.6493, "accuracy": 0.59375, "lr": 7.375545335697085e-07, "epoch": 0.6765714285714286, "percentage": 67.89, "elapsed_time": "0:44:03", "remaining_time": "0:20:50"}
75
+ {"current_steps": 75, "total_steps": 109, "loss": 0.6563, "accuracy": 0.609375, "lr": 7.00644938073904e-07, "epoch": 0.6857142857142857, "percentage": 68.81, "elapsed_time": "0:44:42", "remaining_time": "0:20:16"}
76
+ {"current_steps": 76, "total_steps": 109, "loss": 0.6547, "accuracy": 0.578125, "lr": 6.643993247438348e-07, "epoch": 0.6948571428571428, "percentage": 69.72, "elapsed_time": "0:45:25", "remaining_time": "0:19:43"}
77
+ {"current_steps": 77, "total_steps": 109, "loss": 0.6406, "accuracy": 0.6875, "lr": 6.288478009022447e-07, "epoch": 0.704, "percentage": 70.64, "elapsed_time": "0:46:29", "remaining_time": "0:19:19"}
78
+ {"current_steps": 78, "total_steps": 109, "loss": 0.6392, "accuracy": 0.578125, "lr": 5.940198973282838e-07, "epoch": 0.7131428571428572, "percentage": 71.56, "elapsed_time": "0:47:13", "remaining_time": "0:18:46"}
79
+ {"current_steps": 79, "total_steps": 109, "loss": 0.6458, "accuracy": 0.671875, "lr": 5.599445437278412e-07, "epoch": 0.7222857142857143, "percentage": 72.48, "elapsed_time": "0:47:55", "remaining_time": "0:18:11"}
80
+ {"current_steps": 80, "total_steps": 109, "loss": 0.6593, "accuracy": 0.578125, "lr": 5.266500447031646e-07, "epoch": 0.7314285714285714, "percentage": 73.39, "elapsed_time": "0:48:34", "remaining_time": "0:17:36"}
81
+ {"current_steps": 81, "total_steps": 109, "loss": 0.6561, "accuracy": 0.609375, "lr": 4.941640562417138e-07, "epoch": 0.7405714285714285, "percentage": 74.31, "elapsed_time": "0:49:09", "remaining_time": "0:16:59"}
82
+ {"current_steps": 82, "total_steps": 109, "loss": 0.653, "accuracy": 0.546875, "lr": 4.6251356274379226e-07, "epoch": 0.7497142857142857, "percentage": 75.23, "elapsed_time": "0:49:45", "remaining_time": "0:16:23"}
83
+ {"current_steps": 83, "total_steps": 109, "loss": 0.673, "accuracy": 0.484375, "lr": 4.317248546080218e-07, "epoch": 0.7588571428571429, "percentage": 76.15, "elapsed_time": "0:50:20", "remaining_time": "0:15:46"}
84
+ {"current_steps": 84, "total_steps": 109, "loss": 0.652, "accuracy": 0.65625, "lr": 4.018235063932971e-07, "epoch": 0.768, "percentage": 77.06, "elapsed_time": "0:50:54", "remaining_time": "0:15:09"}
85
+ {"current_steps": 85, "total_steps": 109, "loss": 0.6568, "accuracy": 0.671875, "lr": 3.7283435557534184e-07, "epoch": 0.7771428571428571, "percentage": 77.98, "elapsed_time": "0:51:31", "remaining_time": "0:14:32"}
86
+ {"current_steps": 86, "total_steps": 109, "loss": 0.6509, "accuracy": 0.703125, "lr": 3.447814819155292e-07, "epoch": 0.7862857142857143, "percentage": 78.9, "elapsed_time": "0:52:11", "remaining_time": "0:13:57"}
87
+ {"current_steps": 87, "total_steps": 109, "loss": 0.6331, "accuracy": 0.75, "lr": 3.1768818745908876e-07, "epoch": 0.7954285714285714, "percentage": 79.82, "elapsed_time": "0:52:43", "remaining_time": "0:13:19"}
88
+ {"current_steps": 88, "total_steps": 109, "loss": 0.6472, "accuracy": 0.59375, "lr": 2.915769771793256e-07, "epoch": 0.8045714285714286, "percentage": 80.73, "elapsed_time": "0:53:24", "remaining_time": "0:12:44"}
89
+ {"current_steps": 89, "total_steps": 109, "loss": 0.655, "accuracy": 0.53125, "lr": 2.6646954028392375e-07, "epoch": 0.8137142857142857, "percentage": 81.65, "elapsed_time": "0:54:02", "remaining_time": "0:12:08"}
90
+ {"current_steps": 90, "total_steps": 109, "loss": 0.672, "accuracy": 0.59375, "lr": 2.4238673219886385e-07, "epoch": 0.8228571428571428, "percentage": 82.57, "elapsed_time": "0:54:39", "remaining_time": "0:11:32"}
91
+ {"current_steps": 91, "total_steps": 109, "loss": 0.648, "accuracy": 0.59375, "lr": 2.1934855724491708e-07, "epoch": 0.832, "percentage": 83.49, "elapsed_time": "0:55:16", "remaining_time": "0:10:55"}
92
+ {"current_steps": 92, "total_steps": 109, "loss": 0.645, "accuracy": 0.65625, "lr": 1.9737415202111148e-07, "epoch": 0.8411428571428572, "percentage": 84.4, "elapsed_time": "0:55:55", "remaining_time": "0:10:20"}
93
+ {"current_steps": 93, "total_steps": 109, "loss": 0.6775, "accuracy": 0.5, "lr": 1.764817695089636e-07, "epoch": 0.8502857142857143, "percentage": 85.32, "elapsed_time": "0:56:31", "remaining_time": "0:09:43"}
94
+ {"current_steps": 94, "total_steps": 109, "loss": 0.6424, "accuracy": 0.609375, "lr": 1.566887639106911e-07, "epoch": 0.8594285714285714, "percentage": 86.24, "elapsed_time": "0:57:09", "remaining_time": "0:09:07"}
95
+ {"current_steps": 95, "total_steps": 109, "loss": 0.6691, "accuracy": 0.578125, "lr": 1.380115762339877e-07, "epoch": 0.8685714285714285, "percentage": 87.16, "elapsed_time": "0:57:46", "remaining_time": "0:08:30"}
96
+ {"current_steps": 96, "total_steps": 109, "loss": 0.6571, "accuracy": 0.578125, "lr": 1.204657206353459e-07, "epoch": 0.8777142857142857, "percentage": 88.07, "elapsed_time": "0:58:17", "remaining_time": "0:07:53"}
97
+ {"current_steps": 97, "total_steps": 109, "loss": 0.6473, "accuracy": 0.703125, "lr": 1.0406577153326192e-07, "epoch": 0.8868571428571429, "percentage": 88.99, "elapsed_time": "0:58:55", "remaining_time": "0:07:17"}
98
+ {"current_steps": 98, "total_steps": 109, "loss": 0.6489, "accuracy": 0.671875, "lr": 8.882535150203569e-08, "epoch": 0.896, "percentage": 89.91, "elapsed_time": "0:59:34", "remaining_time": "0:06:41"}
99
+ {"current_steps": 99, "total_steps": 109, "loss": 0.6529, "accuracy": 0.59375, "lr": 7.475711995621387e-08, "epoch": 0.9051428571428571, "percentage": 90.83, "elapsed_time": "1:00:14", "remaining_time": "0:06:05"}
100
+ {"current_steps": 100, "total_steps": 109, "loss": 0.6312, "accuracy": 0.640625, "lr": 6.187276263508168e-08, "epoch": 0.9142857142857143, "percentage": 91.74, "elapsed_time": "1:00:51", "remaining_time": "0:05:28"}
101
+ {"current_steps": 101, "total_steps": 109, "loss": 0.6443, "accuracy": 0.640625, "lr": 5.018298189593368e-08, "epoch": 0.9234285714285714, "percentage": 92.66, "elapsed_time": "1:01:32", "remaining_time": "0:04:52"}
102
+ {"current_steps": 102, "total_steps": 109, "loss": 0.6263, "accuracy": 0.734375, "lr": 3.969748782418991e-08, "epoch": 0.9325714285714286, "percentage": 93.58, "elapsed_time": "1:02:09", "remaining_time": "0:04:15"}
103
+ {"current_steps": 103, "total_steps": 109, "loss": 0.6414, "accuracy": 0.65625, "lr": 3.042499016773881e-08, "epoch": 0.9417142857142857, "percentage": 94.5, "elapsed_time": "1:02:44", "remaining_time": "0:03:39"}
104
+ {"current_steps": 104, "total_steps": 109, "loss": 0.6496, "accuracy": 0.53125, "lr": 2.2373191102207647e-08, "epoch": 0.9508571428571428, "percentage": 95.41, "elapsed_time": "1:03:22", "remaining_time": "0:03:02"}
105
+ {"current_steps": 105, "total_steps": 109, "loss": 0.6416, "accuracy": 0.59375, "lr": 1.5548778833171463e-08, "epoch": 0.96, "percentage": 96.33, "elapsed_time": "1:03:55", "remaining_time": "0:02:26"}
106
+ {"current_steps": 106, "total_steps": 109, "loss": 0.6445, "accuracy": 0.625, "lr": 9.957422040612507e-09, "epoch": 0.9691428571428572, "percentage": 97.25, "elapsed_time": "1:04:39", "remaining_time": "0:01:49"}
107
+ {"current_steps": 107, "total_steps": 109, "loss": 0.6312, "accuracy": 0.765625, "lr": 5.6037651702463e-09, "epoch": 0.9782857142857143, "percentage": 98.17, "elapsed_time": "1:05:21", "remaining_time": "0:01:13"}
108
+ {"current_steps": 108, "total_steps": 109, "loss": 0.6424, "accuracy": 0.65625, "lr": 2.491424575625123e-09, "epoch": 0.9874285714285714, "percentage": 99.08, "elapsed_time": "1:06:02", "remaining_time": "0:00:36"}
109
+ {"current_steps": 109, "total_steps": 109, "loss": 0.6358, "accuracy": 0.640625, "lr": 6.229855142232399e-10, "epoch": 0.9965714285714286, "percentage": 100.0, "elapsed_time": "1:06:46", "remaining_time": "0:00:00"}
110
+ {"current_steps": 109, "total_steps": 109, "epoch": 0.9965714285714286, "percentage": 100.0, "elapsed_time": "1:07:03", "remaining_time": "0:00:00"}
trainer_state.json ADDED
@@ -0,0 +1,1678 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.9965714285714286,
6
+ "eval_steps": 500,
7
+ "global_step": 109,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.009142857142857144,
14
+ "grad_norm": 1.1040480136871338,
15
+ "learning_rate": 3e-06,
16
+ "logits/chosen": 0.8711121678352356,
17
+ "logits/rejected": 0.8816854953765869,
18
+ "logps/chosen": -27.17063331604004,
19
+ "logps/rejected": -30.650341033935547,
20
+ "loss": 0.6931,
21
+ "rewards/accuracies": 0.0,
22
+ "rewards/chosen": 0.0,
23
+ "rewards/margins": 0.0,
24
+ "rewards/rejected": 0.0,
25
+ "step": 1
26
+ },
27
+ {
28
+ "epoch": 0.018285714285714287,
29
+ "grad_norm": 1.0154658555984497,
30
+ "learning_rate": 2.9993770144857767e-06,
31
+ "logits/chosen": 0.9941316843032837,
32
+ "logits/rejected": 0.9719477295875549,
33
+ "logps/chosen": -26.567737579345703,
34
+ "logps/rejected": -26.810808181762695,
35
+ "loss": 0.6929,
36
+ "rewards/accuracies": 0.40625,
37
+ "rewards/chosen": 0.00017506798030808568,
38
+ "rewards/margins": 0.000724022276699543,
39
+ "rewards/rejected": -0.0005489542381837964,
40
+ "step": 2
41
+ },
42
+ {
43
+ "epoch": 0.027428571428571427,
44
+ "grad_norm": 1.1707862615585327,
45
+ "learning_rate": 2.997508575424375e-06,
46
+ "logits/chosen": 0.8969916701316833,
47
+ "logits/rejected": 0.8827911615371704,
48
+ "logps/chosen": -33.664066314697266,
49
+ "logps/rejected": -42.806617736816406,
50
+ "loss": 0.6925,
51
+ "rewards/accuracies": 0.390625,
52
+ "rewards/chosen": -0.005040694959461689,
53
+ "rewards/margins": 0.0014650868251919746,
54
+ "rewards/rejected": -0.006505781318992376,
55
+ "step": 3
56
+ },
57
+ {
58
+ "epoch": 0.036571428571428574,
59
+ "grad_norm": 1.2025189399719238,
60
+ "learning_rate": 2.9943962348297537e-06,
61
+ "logits/chosen": 0.8857066631317139,
62
+ "logits/rejected": 0.8838445544242859,
63
+ "logps/chosen": -32.49725341796875,
64
+ "logps/rejected": -37.86215591430664,
65
+ "loss": 0.6941,
66
+ "rewards/accuracies": 0.421875,
67
+ "rewards/chosen": -0.00038554967613890767,
68
+ "rewards/margins": -0.0015564307104796171,
69
+ "rewards/rejected": 0.0011708810925483704,
70
+ "step": 4
71
+ },
72
+ {
73
+ "epoch": 0.045714285714285714,
74
+ "grad_norm": 1.0835425853729248,
75
+ "learning_rate": 2.9900425779593876e-06,
76
+ "logits/chosen": 0.9785147309303284,
77
+ "logits/rejected": 0.9767488837242126,
78
+ "logps/chosen": -32.34138107299805,
79
+ "logps/rejected": -33.06357955932617,
80
+ "loss": 0.6915,
81
+ "rewards/accuracies": 0.5,
82
+ "rewards/chosen": 0.004742755088955164,
83
+ "rewards/margins": 0.003515874035656452,
84
+ "rewards/rejected": 0.0012268810532987118,
85
+ "step": 5
86
+ },
87
+ {
88
+ "epoch": 0.054857142857142854,
89
+ "grad_norm": 1.1358180046081543,
90
+ "learning_rate": 2.9844512211668286e-06,
91
+ "logits/chosen": 0.975345253944397,
92
+ "logits/rejected": 0.9747197031974792,
93
+ "logps/chosen": -39.865135192871094,
94
+ "logps/rejected": -41.6209716796875,
95
+ "loss": 0.6925,
96
+ "rewards/accuracies": 0.421875,
97
+ "rewards/chosen": -0.0015356692019850016,
98
+ "rewards/margins": 0.0015356941148638725,
99
+ "rewards/rejected": -0.0030713630840182304,
100
+ "step": 6
101
+ },
102
+ {
103
+ "epoch": 0.064,
104
+ "grad_norm": 1.0809563398361206,
105
+ "learning_rate": 2.977626808897792e-06,
106
+ "logits/chosen": 0.9862444400787354,
107
+ "logits/rejected": 0.9724388122558594,
108
+ "logps/chosen": -29.686092376708984,
109
+ "logps/rejected": -31.907615661621094,
110
+ "loss": 0.6891,
111
+ "rewards/accuracies": 0.53125,
112
+ "rewards/chosen": -3.781175473704934e-05,
113
+ "rewards/margins": 0.008373227901756763,
114
+ "rewards/rejected": -0.008411040529608727,
115
+ "step": 7
116
+ },
117
+ {
118
+ "epoch": 0.07314285714285715,
119
+ "grad_norm": 1.0724674463272095,
120
+ "learning_rate": 2.9695750098322613e-06,
121
+ "logits/chosen": 0.8200769424438477,
122
+ "logits/rejected": 0.8287712931632996,
123
+ "logps/chosen": -28.44533920288086,
124
+ "logps/rejected": -31.404376983642578,
125
+ "loss": 0.6949,
126
+ "rewards/accuracies": 0.359375,
127
+ "rewards/chosen": -0.00521398801356554,
128
+ "rewards/margins": -0.003260388970375061,
129
+ "rewards/rejected": -0.0019535990431904793,
130
+ "step": 8
131
+ },
132
+ {
133
+ "epoch": 0.08228571428571428,
134
+ "grad_norm": 1.2545056343078613,
135
+ "learning_rate": 2.9603025121758102e-06,
136
+ "logits/chosen": 0.8948537707328796,
137
+ "logits/rejected": 0.9068803787231445,
138
+ "logps/chosen": -46.04057312011719,
139
+ "logps/rejected": -48.713165283203125,
140
+ "loss": 0.6907,
141
+ "rewards/accuracies": 0.46875,
142
+ "rewards/chosen": -0.0005462196422740817,
143
+ "rewards/margins": 0.005249999463558197,
144
+ "rewards/rejected": -0.0057962192222476006,
145
+ "step": 9
146
+ },
147
+ {
148
+ "epoch": 0.09142857142857143,
149
+ "grad_norm": 1.0415420532226562,
150
+ "learning_rate": 2.9498170181040663e-06,
151
+ "logits/chosen": 0.9541230201721191,
152
+ "logits/rejected": 0.9705378413200378,
153
+ "logps/chosen": -45.06901931762695,
154
+ "logps/rejected": -49.19139099121094,
155
+ "loss": 0.6899,
156
+ "rewards/accuracies": 0.53125,
157
+ "rewards/chosen": 0.008702115155756474,
158
+ "rewards/margins": 0.006677050143480301,
159
+ "rewards/rejected": 0.0020250650122761726,
160
+ "step": 10
161
+ },
162
+ {
163
+ "epoch": 0.10057142857142858,
164
+ "grad_norm": 1.1676886081695557,
165
+ "learning_rate": 2.938127237364918e-06,
166
+ "logits/chosen": 0.986663281917572,
167
+ "logits/rejected": 1.0067094564437866,
168
+ "logps/chosen": -28.37976837158203,
169
+ "logps/rejected": -39.75647735595703,
170
+ "loss": 0.692,
171
+ "rewards/accuracies": 0.484375,
172
+ "rewards/chosen": 0.00539613701403141,
173
+ "rewards/margins": 0.002474667504429817,
174
+ "rewards/rejected": 0.002921469509601593,
175
+ "step": 11
176
+ },
177
+ {
178
+ "epoch": 0.10971428571428571,
179
+ "grad_norm": 1.2175946235656738,
180
+ "learning_rate": 2.925242880043786e-06,
181
+ "logits/chosen": 0.9299088716506958,
182
+ "logits/rejected": 0.929700493812561,
183
+ "logps/chosen": -38.260108947753906,
184
+ "logps/rejected": -28.642169952392578,
185
+ "loss": 0.6886,
186
+ "rewards/accuracies": 0.59375,
187
+ "rewards/chosen": 0.005420793779194355,
188
+ "rewards/margins": 0.009435496293008327,
189
+ "rewards/rejected": -0.0040147025138139725,
190
+ "step": 12
191
+ },
192
+ {
193
+ "epoch": 0.11885714285714286,
194
+ "grad_norm": 1.0878849029541016,
195
+ "learning_rate": 2.911174648497964e-06,
196
+ "logits/chosen": 0.9711499214172363,
197
+ "logits/rejected": 0.9902426600456238,
198
+ "logps/chosen": -32.96807098388672,
199
+ "logps/rejected": -27.930086135864258,
200
+ "loss": 0.6912,
201
+ "rewards/accuracies": 0.46875,
202
+ "rewards/chosen": 0.005747639108449221,
203
+ "rewards/margins": 0.004166706465184689,
204
+ "rewards/rejected": 0.0015809328760951757,
205
+ "step": 13
206
+ },
207
+ {
208
+ "epoch": 0.128,
209
+ "grad_norm": 1.2230732440948486,
210
+ "learning_rate": 2.895934228466738e-06,
211
+ "logits/chosen": 0.939419150352478,
212
+ "logits/rejected": 0.9696264266967773,
213
+ "logps/chosen": -26.001445770263672,
214
+ "logps/rejected": -34.674659729003906,
215
+ "loss": 0.6924,
216
+ "rewards/accuracies": 0.46875,
217
+ "rewards/chosen": 0.004261075984686613,
218
+ "rewards/margins": 0.001839590142481029,
219
+ "rewards/rejected": 0.002421486424282193,
220
+ "step": 14
221
+ },
222
+ {
223
+ "epoch": 0.13714285714285715,
224
+ "grad_norm": 0.9862863421440125,
225
+ "learning_rate": 2.879534279364654e-06,
226
+ "logits/chosen": 0.8570408225059509,
227
+ "logits/rejected": 0.8515439033508301,
228
+ "logps/chosen": -30.134687423706055,
229
+ "logps/rejected": -34.17374801635742,
230
+ "loss": 0.692,
231
+ "rewards/accuracies": 0.515625,
232
+ "rewards/chosen": 0.008894287049770355,
233
+ "rewards/margins": 0.002626563888043165,
234
+ "rewards/rejected": 0.006267723627388477,
235
+ "step": 15
236
+ },
237
+ {
238
+ "epoch": 0.1462857142857143,
239
+ "grad_norm": 1.1977207660675049,
240
+ "learning_rate": 2.8619884237660125e-06,
241
+ "logits/chosen": 0.8890377283096313,
242
+ "logits/rejected": 0.8864951133728027,
243
+ "logps/chosen": -29.793222427368164,
244
+ "logps/rejected": -39.94226837158203,
245
+ "loss": 0.6953,
246
+ "rewards/accuracies": 0.390625,
247
+ "rewards/chosen": 0.009408067911863327,
248
+ "rewards/margins": -0.003934869077056646,
249
+ "rewards/rejected": 0.013342936523258686,
250
+ "step": 16
251
+ },
252
+ {
253
+ "epoch": 0.15542857142857142,
254
+ "grad_norm": 1.0249710083007812,
255
+ "learning_rate": 2.843311236089309e-06,
256
+ "logits/chosen": 1.0601900815963745,
257
+ "logits/rejected": 1.0796374082565308,
258
+ "logps/chosen": -22.465839385986328,
259
+ "logps/rejected": -25.507183074951172,
260
+ "loss": 0.6885,
261
+ "rewards/accuracies": 0.46875,
262
+ "rewards/chosen": 0.007908504456281662,
263
+ "rewards/margins": 0.009687112644314766,
264
+ "rewards/rejected": -0.001778608770109713,
265
+ "step": 17
266
+ },
267
+ {
268
+ "epoch": 0.16457142857142856,
269
+ "grad_norm": 1.154029369354248,
270
+ "learning_rate": 2.8235182304910364e-06,
271
+ "logits/chosen": 0.9637155532836914,
272
+ "logits/rejected": 0.9883713722229004,
273
+ "logps/chosen": -30.487091064453125,
274
+ "logps/rejected": -32.88225555419922,
275
+ "loss": 0.6869,
276
+ "rewards/accuracies": 0.546875,
277
+ "rewards/chosen": 0.002291870303452015,
278
+ "rewards/margins": 0.012858748435974121,
279
+ "rewards/rejected": -0.01056687906384468,
280
+ "step": 18
281
+ },
282
+ {
283
+ "epoch": 0.1737142857142857,
284
+ "grad_norm": 1.0419321060180664,
285
+ "learning_rate": 2.8026258479788888e-06,
286
+ "logits/chosen": 0.8351796865463257,
287
+ "logits/rejected": 0.8506074547767639,
288
+ "logps/chosen": -30.02130889892578,
289
+ "logps/rejected": -32.204925537109375,
290
+ "loss": 0.6882,
291
+ "rewards/accuracies": 0.53125,
292
+ "rewards/chosen": 0.008015453815460205,
293
+ "rewards/margins": 0.010326000861823559,
294
+ "rewards/rejected": -0.0023105472791939974,
295
+ "step": 19
296
+ },
297
+ {
298
+ "epoch": 0.18285714285714286,
299
+ "grad_norm": 1.277442216873169,
300
+ "learning_rate": 2.780651442755083e-06,
301
+ "logits/chosen": 0.9407975077629089,
302
+ "logits/rejected": 0.9294748306274414,
303
+ "logps/chosen": -28.095020294189453,
304
+ "logps/rejected": -40.552581787109375,
305
+ "loss": 0.6856,
306
+ "rewards/accuracies": 0.59375,
307
+ "rewards/chosen": 0.010045948438346386,
308
+ "rewards/margins": 0.015951840206980705,
309
+ "rewards/rejected": -0.005905891768634319,
310
+ "step": 20
311
+ },
312
+ {
313
+ "epoch": 0.192,
314
+ "grad_norm": 1.08894681930542,
315
+ "learning_rate": 2.7576132678011365e-06,
316
+ "logits/chosen": 0.8888410925865173,
317
+ "logits/rejected": 0.8817054033279419,
318
+ "logps/chosen": -35.07438278198242,
319
+ "logps/rejected": -34.59785842895508,
320
+ "loss": 0.6852,
321
+ "rewards/accuracies": 0.578125,
322
+ "rewards/chosen": 0.006834886036813259,
323
+ "rewards/margins": 0.016209105029702187,
324
+ "rewards/rejected": -0.009374218061566353,
325
+ "step": 21
326
+ },
327
+ {
328
+ "epoch": 0.20114285714285715,
329
+ "grad_norm": 1.0777868032455444,
330
+ "learning_rate": 2.7335304597160764e-06,
331
+ "logits/chosen": 0.9650536179542542,
332
+ "logits/rejected": 0.973263680934906,
333
+ "logps/chosen": -35.255348205566406,
334
+ "logps/rejected": -31.095020294189453,
335
+ "loss": 0.6849,
336
+ "rewards/accuracies": 0.484375,
337
+ "rewards/chosen": 0.015151962637901306,
338
+ "rewards/margins": 0.01791677251458168,
339
+ "rewards/rejected": -0.0027648117393255234,
340
+ "step": 22
341
+ },
342
+ {
343
+ "epoch": 0.2102857142857143,
344
+ "grad_norm": 1.1337324380874634,
345
+ "learning_rate": 2.7084230228206746e-06,
346
+ "logits/chosen": 0.9315263628959656,
347
+ "logits/rejected": 0.9389154314994812,
348
+ "logps/chosen": -29.42315673828125,
349
+ "logps/rejected": -39.68217086791992,
350
+ "loss": 0.6857,
351
+ "rewards/accuracies": 0.515625,
352
+ "rewards/chosen": 0.007597707211971283,
353
+ "rewards/margins": 0.015555331483483315,
354
+ "rewards/rejected": -0.007957624271512032,
355
+ "step": 23
356
+ },
357
+ {
358
+ "epoch": 0.21942857142857142,
359
+ "grad_norm": 1.1391428709030151,
360
+ "learning_rate": 2.6823118125409112e-06,
361
+ "logits/chosen": 0.9457556009292603,
362
+ "logits/rejected": 0.9379363059997559,
363
+ "logps/chosen": -33.75120544433594,
364
+ "logps/rejected": -36.649776458740234,
365
+ "loss": 0.6854,
366
+ "rewards/accuracies": 0.484375,
367
+ "rewards/chosen": 0.008110197260975838,
368
+ "rewards/margins": 0.01610742136836052,
369
+ "rewards/rejected": -0.007997222244739532,
370
+ "step": 24
371
+ },
372
+ {
373
+ "epoch": 0.22857142857142856,
374
+ "grad_norm": 1.4407936334609985,
375
+ "learning_rate": 2.6552185180844704e-06,
376
+ "logits/chosen": 0.9158231019973755,
377
+ "logits/rejected": 0.9063650369644165,
378
+ "logps/chosen": -33.36897277832031,
379
+ "logps/rejected": -56.65917205810547,
380
+ "loss": 0.6858,
381
+ "rewards/accuracies": 0.5625,
382
+ "rewards/chosen": 0.0024651410058140755,
383
+ "rewards/margins": 0.015247687697410583,
384
+ "rewards/rejected": -0.012782546691596508,
385
+ "step": 25
386
+ },
387
+ {
388
+ "epoch": 0.2377142857142857,
389
+ "grad_norm": 1.0874488353729248,
390
+ "learning_rate": 2.6271656444246578e-06,
391
+ "logits/chosen": 0.9440329670906067,
392
+ "logits/rejected": 0.9391080737113953,
393
+ "logps/chosen": -23.501907348632812,
394
+ "logps/rejected": -30.94306755065918,
395
+ "loss": 0.6822,
396
+ "rewards/accuracies": 0.59375,
397
+ "rewards/chosen": 0.0028711333870887756,
398
+ "rewards/margins": 0.022652385756373405,
399
+ "rewards/rejected": -0.01978125423192978,
400
+ "step": 26
401
+ },
402
+ {
403
+ "epoch": 0.24685714285714286,
404
+ "grad_norm": 1.2141540050506592,
405
+ "learning_rate": 2.598176493606703e-06,
406
+ "logits/chosen": 0.9432783722877502,
407
+ "logits/rejected": 0.9551030397415161,
408
+ "logps/chosen": -36.34120559692383,
409
+ "logps/rejected": -42.346473693847656,
410
+ "loss": 0.6912,
411
+ "rewards/accuracies": 0.484375,
412
+ "rewards/chosen": 0.0062694246880710125,
413
+ "rewards/margins": 0.004517142660915852,
414
+ "rewards/rejected": 0.0017522834241390228,
415
+ "step": 27
416
+ },
417
+ {
418
+ "epoch": 0.256,
419
+ "grad_norm": 1.0875886678695679,
420
+ "learning_rate": 2.568275145391978e-06,
421
+ "logits/chosen": 0.822512686252594,
422
+ "logits/rejected": 0.7919052839279175,
423
+ "logps/chosen": -40.985992431640625,
424
+ "logps/rejected": -48.13853454589844,
425
+ "loss": 0.6845,
426
+ "rewards/accuracies": 0.625,
427
+ "rewards/chosen": 0.01346580870449543,
428
+ "rewards/margins": 0.01811329834163189,
429
+ "rewards/rejected": -0.004647488705813885,
430
+ "step": 28
431
+ },
432
+ {
433
+ "epoch": 0.2651428571428571,
434
+ "grad_norm": 1.065311074256897,
435
+ "learning_rate": 2.5374864372562077e-06,
436
+ "logits/chosen": 0.9807026386260986,
437
+ "logits/rejected": 0.9859702587127686,
438
+ "logps/chosen": -28.136564254760742,
439
+ "logps/rejected": -33.921974182128906,
440
+ "loss": 0.6784,
441
+ "rewards/accuracies": 0.625,
442
+ "rewards/chosen": -0.0027041807770729065,
443
+ "rewards/margins": 0.03066922351717949,
444
+ "rewards/rejected": -0.033373404294252396,
445
+ "step": 29
446
+ },
447
+ {
448
+ "epoch": 0.2742857142857143,
449
+ "grad_norm": 1.111535906791687,
450
+ "learning_rate": 2.505835943758286e-06,
451
+ "logits/chosen": 1.0877310037612915,
452
+ "logits/rejected": 1.0837771892547607,
453
+ "logps/chosen": -26.631589889526367,
454
+ "logps/rejected": -28.076698303222656,
455
+ "loss": 0.6809,
456
+ "rewards/accuracies": 0.625,
457
+ "rewards/chosen": 0.001617547939531505,
458
+ "rewards/margins": 0.025237493216991425,
459
+ "rewards/rejected": -0.023619946092367172,
460
+ "step": 30
461
+ },
462
+ {
463
+ "epoch": 0.2834285714285714,
464
+ "grad_norm": 1.1810338497161865,
465
+ "learning_rate": 2.4733499552968357e-06,
466
+ "logits/chosen": 1.0420756340026855,
467
+ "logits/rejected": 1.0458323955535889,
468
+ "logps/chosen": -33.388362884521484,
469
+ "logps/rejected": -27.1654052734375,
470
+ "loss": 0.678,
471
+ "rewards/accuracies": 0.53125,
472
+ "rewards/chosen": -0.00019822129979729652,
473
+ "rewards/margins": 0.03170105069875717,
474
+ "rewards/rejected": -0.03189927339553833,
475
+ "step": 31
476
+ },
477
+ {
478
+ "epoch": 0.2925714285714286,
479
+ "grad_norm": 1.3464339971542358,
480
+ "learning_rate": 2.440055456272159e-06,
481
+ "logits/chosen": 1.0149445533752441,
482
+ "logits/rejected": 1.0009658336639404,
483
+ "logps/chosen": -33.061546325683594,
484
+ "logps/rejected": -43.69398498535156,
485
+ "loss": 0.669,
486
+ "rewards/accuracies": 0.71875,
487
+ "rewards/chosen": 0.0036624798085540533,
488
+ "rewards/margins": 0.0504765547811985,
489
+ "rewards/rejected": -0.046814076602458954,
490
+ "step": 32
491
+ },
492
+ {
493
+ "epoch": 0.3017142857142857,
494
+ "grad_norm": 1.1221750974655151,
495
+ "learning_rate": 2.4059801026717166e-06,
496
+ "logits/chosen": 0.925652265548706,
497
+ "logits/rejected": 0.9217759370803833,
498
+ "logps/chosen": -34.813316345214844,
499
+ "logps/rejected": -39.314117431640625,
500
+ "loss": 0.678,
501
+ "rewards/accuracies": 0.640625,
502
+ "rewards/chosen": -0.013261141255497932,
503
+ "rewards/margins": 0.03146106004714966,
504
+ "rewards/rejected": -0.04472219944000244,
505
+ "step": 33
506
+ },
507
+ {
508
+ "epoch": 0.31085714285714283,
509
+ "grad_norm": 1.186403512954712,
510
+ "learning_rate": 2.3711521990977554e-06,
511
+ "logits/chosen": 0.8498350381851196,
512
+ "logits/rejected": 0.8486256003379822,
513
+ "logps/chosen": -36.290618896484375,
514
+ "logps/rejected": -31.55414581298828,
515
+ "loss": 0.6784,
516
+ "rewards/accuracies": 0.5625,
517
+ "rewards/chosen": -0.008366885595023632,
518
+ "rewards/margins": 0.031201709061861038,
519
+ "rewards/rejected": -0.039568595588207245,
520
+ "step": 34
521
+ },
522
+ {
523
+ "epoch": 0.32,
524
+ "grad_norm": 1.168227195739746,
525
+ "learning_rate": 2.3356006752561658e-06,
526
+ "logits/chosen": 0.7546634674072266,
527
+ "logits/rejected": 0.7651799321174622,
528
+ "logps/chosen": -23.884647369384766,
529
+ "logps/rejected": -27.636383056640625,
530
+ "loss": 0.6812,
531
+ "rewards/accuracies": 0.53125,
532
+ "rewards/chosen": -0.008346683345735073,
533
+ "rewards/margins": 0.02539180964231491,
534
+ "rewards/rejected": -0.03373849391937256,
535
+ "step": 35
536
+ },
537
+ {
538
+ "epoch": 0.3291428571428571,
539
+ "grad_norm": 1.1387567520141602,
540
+ "learning_rate": 2.299355061926096e-06,
541
+ "logits/chosen": 0.8960639238357544,
542
+ "logits/rejected": 0.9027845859527588,
543
+ "logps/chosen": -31.159500122070312,
544
+ "logps/rejected": -33.181968688964844,
545
+ "loss": 0.6691,
546
+ "rewards/accuracies": 0.65625,
547
+ "rewards/chosen": -0.0007302656304091215,
548
+ "rewards/margins": 0.050405994057655334,
549
+ "rewards/rejected": -0.051136258989572525,
550
+ "step": 36
551
+ },
552
+ {
553
+ "epoch": 0.3382857142857143,
554
+ "grad_norm": 1.1846139430999756,
555
+ "learning_rate": 2.262445466430292e-06,
556
+ "logits/chosen": 1.0242822170257568,
557
+ "logits/rejected": 1.0171458721160889,
558
+ "logps/chosen": -32.45991516113281,
559
+ "logps/rejected": -33.43197250366211,
560
+ "loss": 0.6801,
561
+ "rewards/accuracies": 0.65625,
562
+ "rewards/chosen": -0.01042199693620205,
563
+ "rewards/margins": 0.027484579011797905,
564
+ "rewards/rejected": -0.03790657967329025,
565
+ "step": 37
566
+ },
567
+ {
568
+ "epoch": 0.3474285714285714,
569
+ "grad_norm": 1.1609152555465698,
570
+ "learning_rate": 2.2249025476265262e-06,
571
+ "logits/chosen": 0.9610393047332764,
572
+ "logits/rejected": 0.9824803471565247,
573
+ "logps/chosen": -37.18042755126953,
574
+ "logps/rejected": -39.24803161621094,
575
+ "loss": 0.6679,
576
+ "rewards/accuracies": 0.609375,
577
+ "rewards/chosen": -0.011728843674063683,
578
+ "rewards/margins": 0.05284466594457626,
579
+ "rewards/rejected": -0.0645735040307045,
580
+ "step": 38
581
+ },
582
+ {
583
+ "epoch": 0.3565714285714286,
584
+ "grad_norm": 1.106920599937439,
585
+ "learning_rate": 2.1867574904409007e-06,
586
+ "logits/chosen": 0.9217398762702942,
587
+ "logits/rejected": 0.9055342674255371,
588
+ "logps/chosen": -28.668115615844727,
589
+ "logps/rejected": -31.50359344482422,
590
+ "loss": 0.6718,
591
+ "rewards/accuracies": 0.703125,
592
+ "rewards/chosen": -0.011347142979502678,
593
+ "rewards/margins": 0.04483164846897125,
594
+ "rewards/rejected": -0.05617879331111908,
595
+ "step": 39
596
+ },
597
+ {
598
+ "epoch": 0.3657142857142857,
599
+ "grad_norm": 1.0243180990219116,
600
+ "learning_rate": 2.1480419799641695e-06,
601
+ "logits/chosen": 0.9214224219322205,
602
+ "logits/rejected": 0.9363438487052917,
603
+ "logps/chosen": -24.622417449951172,
604
+ "logps/rejected": -26.77389144897461,
605
+ "loss": 0.6774,
606
+ "rewards/accuracies": 0.578125,
607
+ "rewards/chosen": -0.013774814084172249,
608
+ "rewards/margins": 0.03302985802292824,
609
+ "rewards/rejected": -0.04680467024445534,
610
+ "step": 40
611
+ },
612
+ {
613
+ "epoch": 0.37485714285714283,
614
+ "grad_norm": 1.196673035621643,
615
+ "learning_rate": 2.1087881751326035e-06,
616
+ "logits/chosen": 0.9006940126419067,
617
+ "logits/rejected": 0.9225171208381653,
618
+ "logps/chosen": -23.879051208496094,
619
+ "logps/rejected": -34.061222076416016,
620
+ "loss": 0.6628,
621
+ "rewards/accuracies": 0.671875,
622
+ "rewards/chosen": -0.01365719921886921,
623
+ "rewards/margins": 0.06352591514587402,
624
+ "rewards/rejected": -0.07718311995267868,
625
+ "step": 41
626
+ },
627
+ {
628
+ "epoch": 0.384,
629
+ "grad_norm": 1.121903657913208,
630
+ "learning_rate": 2.0690286820152535e-06,
631
+ "logits/chosen": 0.8951718211174011,
632
+ "logits/rejected": 0.8833389282226562,
633
+ "logps/chosen": -31.429443359375,
634
+ "logps/rejected": -29.13869857788086,
635
+ "loss": 0.6767,
636
+ "rewards/accuracies": 0.59375,
637
+ "rewards/chosen": -0.019592294469475746,
638
+ "rewards/margins": 0.034957218915224075,
639
+ "rewards/rejected": -0.05454951152205467,
640
+ "step": 42
641
+ },
642
+ {
643
+ "epoch": 0.3931428571428571,
644
+ "grad_norm": 1.1516579389572144,
645
+ "learning_rate": 2.028796526729806e-06,
646
+ "logits/chosen": 0.8912935256958008,
647
+ "logits/rejected": 0.8846753835678101,
648
+ "logps/chosen": -31.40945053100586,
649
+ "logps/rejected": -30.121349334716797,
650
+ "loss": 0.6676,
651
+ "rewards/accuracies": 0.578125,
652
+ "rewards/chosen": -0.015952344983816147,
653
+ "rewards/margins": 0.05502069741487503,
654
+ "rewards/rejected": -0.07097303867340088,
655
+ "step": 43
656
+ },
657
+ {
658
+ "epoch": 0.4022857142857143,
659
+ "grad_norm": 1.307623028755188,
660
+ "learning_rate": 1.9881251280095263e-06,
661
+ "logits/chosen": 0.9057846069335938,
662
+ "logits/rejected": 0.8929940462112427,
663
+ "logps/chosen": -35.15623474121094,
664
+ "logps/rejected": -40.14659881591797,
665
+ "loss": 0.6733,
666
+ "rewards/accuracies": 0.609375,
667
+ "rewards/chosen": -0.01705355755984783,
668
+ "rewards/margins": 0.04369322955608368,
669
+ "rewards/rejected": -0.06074678897857666,
670
+ "step": 44
671
+ },
672
+ {
673
+ "epoch": 0.4114285714285714,
674
+ "grad_norm": 1.1834821701049805,
675
+ "learning_rate": 1.9470482694440755e-06,
676
+ "logits/chosen": 0.9042454957962036,
677
+ "logits/rejected": 0.8868681192398071,
678
+ "logps/chosen": -30.832304000854492,
679
+ "logps/rejected": -30.351964950561523,
680
+ "loss": 0.6639,
681
+ "rewards/accuracies": 0.703125,
682
+ "rewards/chosen": -0.002259607659652829,
683
+ "rewards/margins": 0.061560556292533875,
684
+ "rewards/rejected": -0.06382016837596893,
685
+ "step": 45
686
+ },
687
+ {
688
+ "epoch": 0.4205714285714286,
689
+ "grad_norm": 1.174766182899475,
690
+ "learning_rate": 1.9056000714172617e-06,
691
+ "logits/chosen": 0.9097862243652344,
692
+ "logits/rejected": 0.9001797437667847,
693
+ "logps/chosen": -29.15369415283203,
694
+ "logps/rejected": -29.355531692504883,
695
+ "loss": 0.6574,
696
+ "rewards/accuracies": 0.6875,
697
+ "rewards/chosen": -0.004840398207306862,
698
+ "rewards/margins": 0.07588091492652893,
699
+ "rewards/rejected": -0.08072131872177124,
700
+ "step": 46
701
+ },
702
+ {
703
+ "epoch": 0.4297142857142857,
704
+ "grad_norm": 1.0384345054626465,
705
+ "learning_rate": 1.8638149627650335e-06,
706
+ "logits/chosen": 0.91684889793396,
707
+ "logits/rejected": 0.8941690921783447,
708
+ "logps/chosen": -33.729591369628906,
709
+ "logps/rejected": -36.042083740234375,
710
+ "loss": 0.6709,
711
+ "rewards/accuracies": 0.53125,
712
+ "rewards/chosen": -0.01630518212914467,
713
+ "rewards/margins": 0.048165224492549896,
714
+ "rewards/rejected": -0.06447041034698486,
715
+ "step": 47
716
+ },
717
+ {
718
+ "epoch": 0.43885714285714283,
719
+ "grad_norm": 1.3309643268585205,
720
+ "learning_rate": 1.8217276521772582e-06,
721
+ "logits/chosen": 1.0668003559112549,
722
+ "logits/rejected": 1.0641244649887085,
723
+ "logps/chosen": -24.097837448120117,
724
+ "logps/rejected": -47.95822525024414,
725
+ "loss": 0.6599,
726
+ "rewards/accuracies": 0.59375,
727
+ "rewards/chosen": -0.012660925276577473,
728
+ "rewards/margins": 0.07291898131370544,
729
+ "rewards/rejected": -0.08557990193367004,
730
+ "step": 48
731
+ },
732
+ {
733
+ "epoch": 0.448,
734
+ "grad_norm": 1.223953366279602,
735
+ "learning_rate": 1.7793730993670408e-06,
736
+ "logits/chosen": 0.903070867061615,
737
+ "logits/rejected": 0.9047319889068604,
738
+ "logps/chosen": -42.516998291015625,
739
+ "logps/rejected": -38.09394836425781,
740
+ "loss": 0.6695,
741
+ "rewards/accuracies": 0.5625,
742
+ "rewards/chosen": -0.014069934375584126,
743
+ "rewards/margins": 0.0504281222820282,
744
+ "rewards/rejected": -0.06449805200099945,
745
+ "step": 49
746
+ },
747
+ {
748
+ "epoch": 0.45714285714285713,
749
+ "grad_norm": 1.1348663568496704,
750
+ "learning_rate": 1.736786486031531e-06,
751
+ "logits/chosen": 0.8666382431983948,
752
+ "logits/rejected": 0.8691962957382202,
753
+ "logps/chosen": -31.790327072143555,
754
+ "logps/rejected": -32.83308792114258,
755
+ "loss": 0.6651,
756
+ "rewards/accuracies": 0.65625,
757
+ "rewards/chosen": -0.004812741186469793,
758
+ "rewards/margins": 0.0594930499792099,
759
+ "rewards/rejected": -0.06430579721927643,
760
+ "step": 50
761
+ },
762
+ {
763
+ "epoch": 0.4662857142857143,
764
+ "grad_norm": 1.0801175832748413,
765
+ "learning_rate": 1.6940031866283395e-06,
766
+ "logits/chosen": 0.7520013451576233,
767
+ "logits/rejected": 0.7403222322463989,
768
+ "logps/chosen": -36.27107238769531,
769
+ "logps/rejected": -40.10198974609375,
770
+ "loss": 0.6756,
771
+ "rewards/accuracies": 0.578125,
772
+ "rewards/chosen": -0.028078220784664154,
773
+ "rewards/margins": 0.037895847111940384,
774
+ "rewards/rejected": -0.06597407162189484,
775
+ "step": 51
776
+ },
777
+ {
778
+ "epoch": 0.4754285714285714,
779
+ "grad_norm": 1.1112788915634155,
780
+ "learning_rate": 1.6510587389918377e-06,
781
+ "logits/chosen": 0.8353133797645569,
782
+ "logits/rejected": 0.823691725730896,
783
+ "logps/chosen": -29.32011604309082,
784
+ "logps/rejected": -28.81536865234375,
785
+ "loss": 0.6582,
786
+ "rewards/accuracies": 0.609375,
787
+ "rewards/chosen": -0.02109226956963539,
788
+ "rewards/margins": 0.07520832121372223,
789
+ "rewards/rejected": -0.09630058705806732,
790
+ "step": 52
791
+ },
792
+ {
793
+ "epoch": 0.4845714285714286,
794
+ "grad_norm": 1.1804602146148682,
795
+ "learning_rate": 1.6079888148137507e-06,
796
+ "logits/chosen": 0.8674898147583008,
797
+ "logits/rejected": 0.8565424680709839,
798
+ "logps/chosen": -38.352237701416016,
799
+ "logps/rejected": -42.52186584472656,
800
+ "loss": 0.6594,
801
+ "rewards/accuracies": 0.671875,
802
+ "rewards/chosen": -0.016463566571474075,
803
+ "rewards/margins": 0.07221756875514984,
804
+ "rewards/rejected": -0.08868113905191422,
805
+ "step": 53
806
+ },
807
+ {
808
+ "epoch": 0.4937142857142857,
809
+ "grad_norm": 1.1166019439697266,
810
+ "learning_rate": 1.564829190012561e-06,
811
+ "logits/chosen": 0.9326304197311401,
812
+ "logits/rejected": 0.9363415837287903,
813
+ "logps/chosen": -33.61487579345703,
814
+ "logps/rejected": -34.57853317260742,
815
+ "loss": 0.6668,
816
+ "rewards/accuracies": 0.640625,
817
+ "rewards/chosen": -0.01470315270125866,
818
+ "rewards/margins": 0.057224590331315994,
819
+ "rewards/rejected": -0.0719277411699295,
820
+ "step": 54
821
+ },
822
+ {
823
+ "epoch": 0.5028571428571429,
824
+ "grad_norm": 1.1633715629577637,
825
+ "learning_rate": 1.521615715016336e-06,
826
+ "logits/chosen": 0.8422874212265015,
827
+ "logits/rejected": 0.8437594175338745,
828
+ "logps/chosen": -30.44858741760254,
829
+ "logps/rejected": -28.049278259277344,
830
+ "loss": 0.6678,
831
+ "rewards/accuracies": 0.578125,
832
+ "rewards/chosen": -0.01843036524951458,
833
+ "rewards/margins": 0.05477701127529144,
834
+ "rewards/rejected": -0.07320737838745117,
835
+ "step": 55
836
+ },
837
+ {
838
+ "epoch": 0.512,
839
+ "grad_norm": 1.0950512886047363,
840
+ "learning_rate": 1.4783842849836645e-06,
841
+ "logits/chosen": 0.8825952410697937,
842
+ "logits/rejected": 0.8785406351089478,
843
+ "logps/chosen": -39.51524353027344,
844
+ "logps/rejected": -46.94993591308594,
845
+ "loss": 0.6781,
846
+ "rewards/accuracies": 0.546875,
847
+ "rewards/chosen": -0.05598952993750572,
848
+ "rewards/margins": 0.0334899015724659,
849
+ "rewards/rejected": -0.08947943896055222,
850
+ "step": 56
851
+ },
852
+ {
853
+ "epoch": 0.5211428571428571,
854
+ "grad_norm": 1.1004191637039185,
855
+ "learning_rate": 1.435170809987439e-06,
856
+ "logits/chosen": 0.9405664801597595,
857
+ "logits/rejected": 0.9432823061943054,
858
+ "logps/chosen": -31.875579833984375,
859
+ "logps/rejected": -35.30060577392578,
860
+ "loss": 0.6551,
861
+ "rewards/accuracies": 0.59375,
862
+ "rewards/chosen": -0.0031816833652555943,
863
+ "rewards/margins": 0.08194887638092041,
864
+ "rewards/rejected": -0.08513056486845016,
865
+ "step": 57
866
+ },
867
+ {
868
+ "epoch": 0.5302857142857142,
869
+ "grad_norm": 1.136016845703125,
870
+ "learning_rate": 1.3920111851862494e-06,
871
+ "logits/chosen": 0.9379781484603882,
872
+ "logits/rejected": 0.9405778050422668,
873
+ "logps/chosen": -23.619964599609375,
874
+ "logps/rejected": -33.969844818115234,
875
+ "loss": 0.6596,
876
+ "rewards/accuracies": 0.703125,
877
+ "rewards/chosen": -0.022212259471416473,
878
+ "rewards/margins": 0.07263332605361938,
879
+ "rewards/rejected": -0.09484558552503586,
880
+ "step": 58
881
+ },
882
+ {
883
+ "epoch": 0.5394285714285715,
884
+ "grad_norm": 1.1601372957229614,
885
+ "learning_rate": 1.3489412610081626e-06,
886
+ "logits/chosen": 1.0101261138916016,
887
+ "logits/rejected": 1.010301113128662,
888
+ "logps/chosen": -33.6581916809082,
889
+ "logps/rejected": -36.06288528442383,
890
+ "loss": 0.6581,
891
+ "rewards/accuracies": 0.6875,
892
+ "rewards/chosen": -0.009537904523313046,
893
+ "rewards/margins": 0.07497294992208481,
894
+ "rewards/rejected": -0.08451086282730103,
895
+ "step": 59
896
+ },
897
+ {
898
+ "epoch": 0.5485714285714286,
899
+ "grad_norm": 1.2016857862472534,
900
+ "learning_rate": 1.3059968133716607e-06,
901
+ "logits/chosen": 0.971367597579956,
902
+ "logits/rejected": 0.9774487018585205,
903
+ "logps/chosen": -37.79004669189453,
904
+ "logps/rejected": -37.27052688598633,
905
+ "loss": 0.6533,
906
+ "rewards/accuracies": 0.671875,
907
+ "rewards/chosen": -0.04108016565442085,
908
+ "rewards/margins": 0.08637972176074982,
909
+ "rewards/rejected": -0.12745988368988037,
910
+ "step": 60
911
+ },
912
+ {
913
+ "epoch": 0.5577142857142857,
914
+ "grad_norm": 1.0725586414337158,
915
+ "learning_rate": 1.2632135139684691e-06,
916
+ "logits/chosen": 0.9029541015625,
917
+ "logits/rejected": 0.8947008848190308,
918
+ "logps/chosen": -35.35030746459961,
919
+ "logps/rejected": -34.78730010986328,
920
+ "loss": 0.6609,
921
+ "rewards/accuracies": 0.640625,
922
+ "rewards/chosen": -0.029115652665495872,
923
+ "rewards/margins": 0.07002345472574234,
924
+ "rewards/rejected": -0.09913910925388336,
925
+ "step": 61
926
+ },
927
+ {
928
+ "epoch": 0.5668571428571428,
929
+ "grad_norm": 1.195502758026123,
930
+ "learning_rate": 1.2206269006329595e-06,
931
+ "logits/chosen": 0.7851489782333374,
932
+ "logits/rejected": 0.7897036075592041,
933
+ "logps/chosen": -27.39234161376953,
934
+ "logps/rejected": -36.095550537109375,
935
+ "loss": 0.6653,
936
+ "rewards/accuracies": 0.515625,
937
+ "rewards/chosen": -0.02338983304798603,
938
+ "rewards/margins": 0.0618324838578701,
939
+ "rewards/rejected": -0.08522231131792068,
940
+ "step": 62
941
+ },
942
+ {
943
+ "epoch": 0.576,
944
+ "grad_norm": 1.2673423290252686,
945
+ "learning_rate": 1.178272347822742e-06,
946
+ "logits/chosen": 0.7174267768859863,
947
+ "logits/rejected": 0.7249353528022766,
948
+ "logps/chosen": -32.0921745300293,
949
+ "logps/rejected": -28.83350372314453,
950
+ "loss": 0.6769,
951
+ "rewards/accuracies": 0.53125,
952
+ "rewards/chosen": -0.03513571247458458,
953
+ "rewards/margins": 0.03682096675038338,
954
+ "rewards/rejected": -0.07195667922496796,
955
+ "step": 63
956
+ },
957
+ {
958
+ "epoch": 0.5851428571428572,
959
+ "grad_norm": 1.2396243810653687,
960
+ "learning_rate": 1.1361850372349668e-06,
961
+ "logits/chosen": 0.831747829914093,
962
+ "logits/rejected": 0.8503408432006836,
963
+ "logps/chosen": -39.849281311035156,
964
+ "logps/rejected": -44.27671813964844,
965
+ "loss": 0.6602,
966
+ "rewards/accuracies": 0.65625,
967
+ "rewards/chosen": -0.04290902987122536,
968
+ "rewards/margins": 0.07199851423501968,
969
+ "rewards/rejected": -0.11490754783153534,
970
+ "step": 64
971
+ },
972
+ {
973
+ "epoch": 0.5942857142857143,
974
+ "grad_norm": 1.2452788352966309,
975
+ "learning_rate": 1.0943999285827381e-06,
976
+ "logits/chosen": 0.9620881676673889,
977
+ "logits/rejected": 0.9350689053535461,
978
+ "logps/chosen": -49.520164489746094,
979
+ "logps/rejected": -56.181514739990234,
980
+ "loss": 0.6563,
981
+ "rewards/accuracies": 0.609375,
982
+ "rewards/chosen": -0.060409560799598694,
983
+ "rewards/margins": 0.08371107280254364,
984
+ "rewards/rejected": -0.14412063360214233,
985
+ "step": 65
986
+ },
987
+ {
988
+ "epoch": 0.6034285714285714,
989
+ "grad_norm": 1.1997966766357422,
990
+ "learning_rate": 1.0529517305559246e-06,
991
+ "logits/chosen": 1.0483283996582031,
992
+ "logits/rejected": 1.0370995998382568,
993
+ "logps/chosen": -32.0396614074707,
994
+ "logps/rejected": -29.256633758544922,
995
+ "loss": 0.6761,
996
+ "rewards/accuracies": 0.484375,
997
+ "rewards/chosen": -0.0514901764690876,
998
+ "rewards/margins": 0.039970673620700836,
999
+ "rewards/rejected": -0.09146084636449814,
1000
+ "step": 66
1001
+ },
1002
+ {
1003
+ "epoch": 0.6125714285714285,
1004
+ "grad_norm": 1.1910446882247925,
1005
+ "learning_rate": 1.0118748719904738e-06,
1006
+ "logits/chosen": 0.8053833246231079,
1007
+ "logits/rejected": 0.7981715202331543,
1008
+ "logps/chosen": -23.898334503173828,
1009
+ "logps/rejected": -35.31452560424805,
1010
+ "loss": 0.6571,
1011
+ "rewards/accuracies": 0.59375,
1012
+ "rewards/chosen": -0.005432231351733208,
1013
+ "rewards/margins": 0.0785951018333435,
1014
+ "rewards/rejected": -0.08402733504772186,
1015
+ "step": 67
1016
+ },
1017
+ {
1018
+ "epoch": 0.6217142857142857,
1019
+ "grad_norm": 1.074831485748291,
1020
+ "learning_rate": 9.712034732701943e-07,
1021
+ "logits/chosen": 0.9318118691444397,
1022
+ "logits/rejected": 0.9297177791595459,
1023
+ "logps/chosen": -31.318601608276367,
1024
+ "logps/rejected": -31.307851791381836,
1025
+ "loss": 0.6603,
1026
+ "rewards/accuracies": 0.640625,
1027
+ "rewards/chosen": -0.02508377842605114,
1028
+ "rewards/margins": 0.07146618515253067,
1029
+ "rewards/rejected": -0.09654996544122696,
1030
+ "step": 68
1031
+ },
1032
+ {
1033
+ "epoch": 0.6308571428571429,
1034
+ "grad_norm": 1.105151653289795,
1035
+ "learning_rate": 9.309713179847465e-07,
1036
+ "logits/chosen": 0.918438732624054,
1037
+ "logits/rejected": 0.9226514101028442,
1038
+ "logps/chosen": -23.845657348632812,
1039
+ "logps/rejected": -28.43765640258789,
1040
+ "loss": 0.6431,
1041
+ "rewards/accuracies": 0.65625,
1042
+ "rewards/chosen": -0.04022161290049553,
1043
+ "rewards/margins": 0.11048796772956848,
1044
+ "rewards/rejected": -0.15070956945419312,
1045
+ "step": 69
1046
+ },
1047
+ {
1048
+ "epoch": 0.64,
1049
+ "grad_norm": 1.0338951349258423,
1050
+ "learning_rate": 8.912118248673967e-07,
1051
+ "logits/chosen": 0.9020978212356567,
1052
+ "logits/rejected": 0.9130126237869263,
1053
+ "logps/chosen": -34.36376953125,
1054
+ "logps/rejected": -40.04608917236328,
1055
+ "loss": 0.6577,
1056
+ "rewards/accuracies": 0.609375,
1057
+ "rewards/chosen": -0.034921795129776,
1058
+ "rewards/margins": 0.07790660858154297,
1059
+ "rewards/rejected": -0.11282840371131897,
1060
+ "step": 70
1061
+ },
1062
+ {
1063
+ "epoch": 0.6491428571428571,
1064
+ "grad_norm": 1.1513733863830566,
1065
+ "learning_rate": 8.519580200358309e-07,
1066
+ "logits/chosen": 0.82832932472229,
1067
+ "logits/rejected": 0.8153083324432373,
1068
+ "logps/chosen": -25.66590118408203,
1069
+ "logps/rejected": -29.915067672729492,
1070
+ "loss": 0.6637,
1071
+ "rewards/accuracies": 0.640625,
1072
+ "rewards/chosen": -0.021308735013008118,
1073
+ "rewards/margins": 0.06582164764404297,
1074
+ "rewards/rejected": -0.08713038265705109,
1075
+ "step": 71
1076
+ },
1077
+ {
1078
+ "epoch": 0.6582857142857143,
1079
+ "grad_norm": 1.2049312591552734,
1080
+ "learning_rate": 8.132425095591e-07,
1081
+ "logits/chosen": 0.8295162916183472,
1082
+ "logits/rejected": 0.8276829719543457,
1083
+ "logps/chosen": -33.05087661743164,
1084
+ "logps/rejected": -44.727622985839844,
1085
+ "loss": 0.634,
1086
+ "rewards/accuracies": 0.640625,
1087
+ "rewards/chosen": -0.032567959278821945,
1088
+ "rewards/margins": 0.13001564145088196,
1089
+ "rewards/rejected": -0.162583589553833,
1090
+ "step": 72
1091
+ },
1092
+ {
1093
+ "epoch": 0.6674285714285715,
1094
+ "grad_norm": 1.2066823244094849,
1095
+ "learning_rate": 7.750974523734742e-07,
1096
+ "logits/chosen": 0.8979049921035767,
1097
+ "logits/rejected": 0.8785669803619385,
1098
+ "logps/chosen": -34.0865364074707,
1099
+ "logps/rejected": -35.59037780761719,
1100
+ "loss": 0.644,
1101
+ "rewards/accuracies": 0.65625,
1102
+ "rewards/chosen": -0.019061893224716187,
1103
+ "rewards/margins": 0.1093374565243721,
1104
+ "rewards/rejected": -0.1283993422985077,
1105
+ "step": 73
1106
+ },
1107
+ {
1108
+ "epoch": 0.6765714285714286,
1109
+ "grad_norm": 1.0506508350372314,
1110
+ "learning_rate": 7.375545335697085e-07,
1111
+ "logits/chosen": 0.9241410493850708,
1112
+ "logits/rejected": 0.920921802520752,
1113
+ "logps/chosen": -32.638404846191406,
1114
+ "logps/rejected": -38.84662628173828,
1115
+ "loss": 0.6493,
1116
+ "rewards/accuracies": 0.59375,
1117
+ "rewards/chosen": -0.02338055893778801,
1118
+ "rewards/margins": 0.09959946572780609,
1119
+ "rewards/rejected": -0.1229800283908844,
1120
+ "step": 74
1121
+ },
1122
+ {
1123
+ "epoch": 0.6857142857142857,
1124
+ "grad_norm": 1.202179193496704,
1125
+ "learning_rate": 7.00644938073904e-07,
1126
+ "logits/chosen": 0.7623097896575928,
1127
+ "logits/rejected": 0.7594358921051025,
1128
+ "logps/chosen": -28.533832550048828,
1129
+ "logps/rejected": -31.006547927856445,
1130
+ "loss": 0.6563,
1131
+ "rewards/accuracies": 0.609375,
1132
+ "rewards/chosen": -0.03972703963518143,
1133
+ "rewards/margins": 0.08638659119606018,
1134
+ "rewards/rejected": -0.126113623380661,
1135
+ "step": 75
1136
+ },
1137
+ {
1138
+ "epoch": 0.6948571428571428,
1139
+ "grad_norm": 1.273539423942566,
1140
+ "learning_rate": 6.643993247438348e-07,
1141
+ "logits/chosen": 0.8061511516571045,
1142
+ "logits/rejected": 0.7985050678253174,
1143
+ "logps/chosen": -29.2039794921875,
1144
+ "logps/rejected": -39.719722747802734,
1145
+ "loss": 0.6547,
1146
+ "rewards/accuracies": 0.578125,
1147
+ "rewards/chosen": -0.04996215179562569,
1148
+ "rewards/margins": 0.09320597350597382,
1149
+ "rewards/rejected": -0.1431681215763092,
1150
+ "step": 76
1151
+ },
1152
+ {
1153
+ "epoch": 0.704,
1154
+ "grad_norm": 1.1269882917404175,
1155
+ "learning_rate": 6.288478009022447e-07,
1156
+ "logits/chosen": 0.9213614463806152,
1157
+ "logits/rejected": 0.9230271577835083,
1158
+ "logps/chosen": -27.4664306640625,
1159
+ "logps/rejected": -35.64370346069336,
1160
+ "loss": 0.6406,
1161
+ "rewards/accuracies": 0.6875,
1162
+ "rewards/chosen": -0.003397151827812195,
1163
+ "rewards/margins": 0.11498977243900299,
1164
+ "rewards/rejected": -0.11838693171739578,
1165
+ "step": 77
1166
+ },
1167
+ {
1168
+ "epoch": 0.7131428571428572,
1169
+ "grad_norm": 1.0317667722702026,
1170
+ "learning_rate": 5.940198973282838e-07,
1171
+ "logits/chosen": 0.9529409408569336,
1172
+ "logits/rejected": 0.9624215364456177,
1173
+ "logps/chosen": -37.00543975830078,
1174
+ "logps/rejected": -40.666053771972656,
1175
+ "loss": 0.6392,
1176
+ "rewards/accuracies": 0.578125,
1177
+ "rewards/chosen": -0.031157776713371277,
1178
+ "rewards/margins": 0.12109382450580597,
1179
+ "rewards/rejected": -0.15225161612033844,
1180
+ "step": 78
1181
+ },
1182
+ {
1183
+ "epoch": 0.7222857142857143,
1184
+ "grad_norm": 1.0846863985061646,
1185
+ "learning_rate": 5.599445437278412e-07,
1186
+ "logits/chosen": 0.8971866369247437,
1187
+ "logits/rejected": 0.90433269739151,
1188
+ "logps/chosen": -32.80634307861328,
1189
+ "logps/rejected": -35.644432067871094,
1190
+ "loss": 0.6458,
1191
+ "rewards/accuracies": 0.671875,
1192
+ "rewards/chosen": -0.04158291965723038,
1193
+ "rewards/margins": 0.1047164648771286,
1194
+ "rewards/rejected": -0.14629937708377838,
1195
+ "step": 79
1196
+ },
1197
+ {
1198
+ "epoch": 0.7314285714285714,
1199
+ "grad_norm": 1.0898044109344482,
1200
+ "learning_rate": 5.266500447031646e-07,
1201
+ "logits/chosen": 0.7604113817214966,
1202
+ "logits/rejected": 0.7796691656112671,
1203
+ "logps/chosen": -25.942516326904297,
1204
+ "logps/rejected": -27.527151107788086,
1205
+ "loss": 0.6593,
1206
+ "rewards/accuracies": 0.578125,
1207
+ "rewards/chosen": -0.02164902538061142,
1208
+ "rewards/margins": 0.07325037568807602,
1209
+ "rewards/rejected": -0.09489940851926804,
1210
+ "step": 80
1211
+ },
1212
+ {
1213
+ "epoch": 0.7405714285714285,
1214
+ "grad_norm": 1.239574909210205,
1215
+ "learning_rate": 4.941640562417138e-07,
1216
+ "logits/chosen": 0.9362255334854126,
1217
+ "logits/rejected": 0.932297945022583,
1218
+ "logps/chosen": -34.35175323486328,
1219
+ "logps/rejected": -27.48154067993164,
1220
+ "loss": 0.6561,
1221
+ "rewards/accuracies": 0.609375,
1222
+ "rewards/chosen": -0.04776213690638542,
1223
+ "rewards/margins": 0.08472548425197601,
1224
+ "rewards/rejected": -0.13248762488365173,
1225
+ "step": 81
1226
+ },
1227
+ {
1228
+ "epoch": 0.7497142857142857,
1229
+ "grad_norm": 1.1007484197616577,
1230
+ "learning_rate": 4.6251356274379226e-07,
1231
+ "logits/chosen": 0.8503745794296265,
1232
+ "logits/rejected": 0.8455989360809326,
1233
+ "logps/chosen": -27.93758201599121,
1234
+ "logps/rejected": -32.705787658691406,
1235
+ "loss": 0.653,
1236
+ "rewards/accuracies": 0.546875,
1237
+ "rewards/chosen": -0.04464199021458626,
1238
+ "rewards/margins": 0.09001848101615906,
1239
+ "rewards/rejected": -0.13466046750545502,
1240
+ "step": 82
1241
+ },
1242
+ {
1243
+ "epoch": 0.7588571428571429,
1244
+ "grad_norm": 1.1005889177322388,
1245
+ "learning_rate": 4.317248546080218e-07,
1246
+ "logits/chosen": 1.045855164527893,
1247
+ "logits/rejected": 1.0475983619689941,
1248
+ "logps/chosen": -22.328994750976562,
1249
+ "logps/rejected": -27.511640548706055,
1250
+ "loss": 0.673,
1251
+ "rewards/accuracies": 0.484375,
1252
+ "rewards/chosen": -0.06321336328983307,
1253
+ "rewards/margins": 0.04760873317718506,
1254
+ "rewards/rejected": -0.11082210391759872,
1255
+ "step": 83
1256
+ },
1257
+ {
1258
+ "epoch": 0.768,
1259
+ "grad_norm": 1.2725178003311157,
1260
+ "learning_rate": 4.018235063932971e-07,
1261
+ "logits/chosen": 0.89765465259552,
1262
+ "logits/rejected": 0.906400203704834,
1263
+ "logps/chosen": -25.914106369018555,
1264
+ "logps/rejected": -38.734458923339844,
1265
+ "loss": 0.652,
1266
+ "rewards/accuracies": 0.65625,
1267
+ "rewards/chosen": -0.038806647062301636,
1268
+ "rewards/margins": 0.09409624338150024,
1269
+ "rewards/rejected": -0.13290289044380188,
1270
+ "step": 84
1271
+ },
1272
+ {
1273
+ "epoch": 0.7771428571428571,
1274
+ "grad_norm": 1.1279877424240112,
1275
+ "learning_rate": 3.7283435557534184e-07,
1276
+ "logits/chosen": 0.9400598406791687,
1277
+ "logits/rejected": 0.9207496047019958,
1278
+ "logps/chosen": -32.89602279663086,
1279
+ "logps/rejected": -35.64267349243164,
1280
+ "loss": 0.6568,
1281
+ "rewards/accuracies": 0.671875,
1282
+ "rewards/chosen": -0.02504522167146206,
1283
+ "rewards/margins": 0.0803045779466629,
1284
+ "rewards/rejected": -0.10534979403018951,
1285
+ "step": 85
1286
+ },
1287
+ {
1288
+ "epoch": 0.7862857142857143,
1289
+ "grad_norm": 1.2422683238983154,
1290
+ "learning_rate": 3.447814819155292e-07,
1291
+ "logits/chosen": 0.8574281334877014,
1292
+ "logits/rejected": 0.8850507140159607,
1293
+ "logps/chosen": -37.92842102050781,
1294
+ "logps/rejected": -47.26014709472656,
1295
+ "loss": 0.6509,
1296
+ "rewards/accuracies": 0.703125,
1297
+ "rewards/chosen": -0.08207320421934128,
1298
+ "rewards/margins": 0.09845928102731705,
1299
+ "rewards/rejected": -0.18053250014781952,
1300
+ "step": 86
1301
+ },
1302
+ {
1303
+ "epoch": 0.7954285714285714,
1304
+ "grad_norm": 1.1461213827133179,
1305
+ "learning_rate": 3.1768818745908876e-07,
1306
+ "logits/chosen": 0.9730187654495239,
1307
+ "logits/rejected": 0.9857529997825623,
1308
+ "logps/chosen": -26.380840301513672,
1309
+ "logps/rejected": -34.13022232055664,
1310
+ "loss": 0.6331,
1311
+ "rewards/accuracies": 0.75,
1312
+ "rewards/chosen": -0.023678911849856377,
1313
+ "rewards/margins": 0.13185349106788635,
1314
+ "rewards/rejected": -0.15553238987922668,
1315
+ "step": 87
1316
+ },
1317
+ {
1318
+ "epoch": 0.8045714285714286,
1319
+ "grad_norm": 1.3225375413894653,
1320
+ "learning_rate": 2.915769771793256e-07,
1321
+ "logits/chosen": 0.8969765901565552,
1322
+ "logits/rejected": 0.8809738159179688,
1323
+ "logps/chosen": -37.93681335449219,
1324
+ "logps/rejected": -36.626155853271484,
1325
+ "loss": 0.6472,
1326
+ "rewards/accuracies": 0.59375,
1327
+ "rewards/chosen": -0.0582134909927845,
1328
+ "rewards/margins": 0.11072604358196259,
1329
+ "rewards/rejected": -0.1689395308494568,
1330
+ "step": 88
1331
+ },
1332
+ {
1333
+ "epoch": 0.8137142857142857,
1334
+ "grad_norm": 1.2331998348236084,
1335
+ "learning_rate": 2.6646954028392375e-07,
1336
+ "logits/chosen": 0.998784065246582,
1337
+ "logits/rejected": 0.9829287528991699,
1338
+ "logps/chosen": -29.3090877532959,
1339
+ "logps/rejected": -38.052547454833984,
1340
+ "loss": 0.655,
1341
+ "rewards/accuracies": 0.53125,
1342
+ "rewards/chosen": -0.01403533760458231,
1343
+ "rewards/margins": 0.09417364001274109,
1344
+ "rewards/rejected": -0.10820896923542023,
1345
+ "step": 89
1346
+ },
1347
+ {
1348
+ "epoch": 0.8228571428571428,
1349
+ "grad_norm": 1.8621482849121094,
1350
+ "learning_rate": 2.4238673219886385e-07,
1351
+ "logits/chosen": 0.9711941480636597,
1352
+ "logits/rejected": 0.9710557460784912,
1353
+ "logps/chosen": -39.98334884643555,
1354
+ "logps/rejected": -37.13236999511719,
1355
+ "loss": 0.672,
1356
+ "rewards/accuracies": 0.59375,
1357
+ "rewards/chosen": -0.09914975613355637,
1358
+ "rewards/margins": 0.06683678925037384,
1359
+ "rewards/rejected": -0.1659865379333496,
1360
+ "step": 90
1361
+ },
1362
+ {
1363
+ "epoch": 0.832,
1364
+ "grad_norm": 1.3211747407913208,
1365
+ "learning_rate": 2.1934855724491708e-07,
1366
+ "logits/chosen": 0.9504058957099915,
1367
+ "logits/rejected": 0.9442592263221741,
1368
+ "logps/chosen": -46.214195251464844,
1369
+ "logps/rejected": -47.317787170410156,
1370
+ "loss": 0.648,
1371
+ "rewards/accuracies": 0.59375,
1372
+ "rewards/chosen": -0.050487205386161804,
1373
+ "rewards/margins": 0.10601682215929031,
1374
+ "rewards/rejected": -0.15650403499603271,
1375
+ "step": 91
1376
+ },
1377
+ {
1378
+ "epoch": 0.8411428571428572,
1379
+ "grad_norm": 1.1896257400512695,
1380
+ "learning_rate": 1.9737415202111148e-07,
1381
+ "logits/chosen": 0.8587050437927246,
1382
+ "logits/rejected": 0.8565249443054199,
1383
+ "logps/chosen": -30.345252990722656,
1384
+ "logps/rejected": -31.947856903076172,
1385
+ "loss": 0.645,
1386
+ "rewards/accuracies": 0.65625,
1387
+ "rewards/chosen": -0.03519854322075844,
1388
+ "rewards/margins": 0.10934263467788696,
1389
+ "rewards/rejected": -0.1445411741733551,
1390
+ "step": 92
1391
+ },
1392
+ {
1393
+ "epoch": 0.8502857142857143,
1394
+ "grad_norm": 1.2513773441314697,
1395
+ "learning_rate": 1.764817695089636e-07,
1396
+ "logits/chosen": 0.9851288795471191,
1397
+ "logits/rejected": 0.9660552144050598,
1398
+ "logps/chosen": -24.57895278930664,
1399
+ "logps/rejected": -31.975099563598633,
1400
+ "loss": 0.6775,
1401
+ "rewards/accuracies": 0.5,
1402
+ "rewards/chosen": -0.05439078062772751,
1403
+ "rewards/margins": 0.04185381531715393,
1404
+ "rewards/rejected": -0.09624460339546204,
1405
+ "step": 93
1406
+ },
1407
+ {
1408
+ "epoch": 0.8594285714285714,
1409
+ "grad_norm": 1.108100175857544,
1410
+ "learning_rate": 1.566887639106911e-07,
1411
+ "logits/chosen": 1.0018175840377808,
1412
+ "logits/rejected": 0.9973495006561279,
1413
+ "logps/chosen": -29.25608253479004,
1414
+ "logps/rejected": -33.57933044433594,
1415
+ "loss": 0.6424,
1416
+ "rewards/accuracies": 0.609375,
1417
+ "rewards/chosen": -0.06291130185127258,
1418
+ "rewards/margins": 0.11450622230768204,
1419
+ "rewards/rejected": -0.17741751670837402,
1420
+ "step": 94
1421
+ },
1422
+ {
1423
+ "epoch": 0.8685714285714285,
1424
+ "grad_norm": 1.4841970205307007,
1425
+ "learning_rate": 1.380115762339877e-07,
1426
+ "logits/chosen": 0.9064180254936218,
1427
+ "logits/rejected": 0.9211479425430298,
1428
+ "logps/chosen": -40.043785095214844,
1429
+ "logps/rejected": -28.14181900024414,
1430
+ "loss": 0.6691,
1431
+ "rewards/accuracies": 0.578125,
1432
+ "rewards/chosen": -0.06435255706310272,
1433
+ "rewards/margins": 0.06316434592008591,
1434
+ "rewards/rejected": -0.12751691043376923,
1435
+ "step": 95
1436
+ },
1437
+ {
1438
+ "epoch": 0.8777142857142857,
1439
+ "grad_norm": 1.2547520399093628,
1440
+ "learning_rate": 1.204657206353459e-07,
1441
+ "logits/chosen": 1.0230824947357178,
1442
+ "logits/rejected": 1.0377960205078125,
1443
+ "logps/chosen": -28.500858306884766,
1444
+ "logps/rejected": -37.0507698059082,
1445
+ "loss": 0.6571,
1446
+ "rewards/accuracies": 0.578125,
1447
+ "rewards/chosen": -0.06484489887952805,
1448
+ "rewards/margins": 0.08328418433666229,
1449
+ "rewards/rejected": -0.14812909066677094,
1450
+ "step": 96
1451
+ },
1452
+ {
1453
+ "epoch": 0.8868571428571429,
1454
+ "grad_norm": 1.3271366357803345,
1455
+ "learning_rate": 1.0406577153326192e-07,
1456
+ "logits/chosen": 0.9916608333587646,
1457
+ "logits/rejected": 0.9704598188400269,
1458
+ "logps/chosen": -38.31233596801758,
1459
+ "logps/rejected": -33.428382873535156,
1460
+ "loss": 0.6473,
1461
+ "rewards/accuracies": 0.703125,
1462
+ "rewards/chosen": -0.02713530696928501,
1463
+ "rewards/margins": 0.1030292958021164,
1464
+ "rewards/rejected": -0.13016459345817566,
1465
+ "step": 97
1466
+ },
1467
+ {
1468
+ "epoch": 0.896,
1469
+ "grad_norm": 1.2687546014785767,
1470
+ "learning_rate": 8.882535150203569e-08,
1471
+ "logits/chosen": 0.8363878726959229,
1472
+ "logits/rejected": 0.817225456237793,
1473
+ "logps/chosen": -36.819740295410156,
1474
+ "logps/rejected": -34.1190299987793,
1475
+ "loss": 0.6489,
1476
+ "rewards/accuracies": 0.671875,
1477
+ "rewards/chosen": -0.04340491071343422,
1478
+ "rewards/margins": 0.09947842359542847,
1479
+ "rewards/rejected": -0.14288334548473358,
1480
+ "step": 98
1481
+ },
1482
+ {
1483
+ "epoch": 0.9051428571428571,
1484
+ "grad_norm": 1.2343368530273438,
1485
+ "learning_rate": 7.475711995621387e-08,
1486
+ "logits/chosen": 0.9504796266555786,
1487
+ "logits/rejected": 0.9603639841079712,
1488
+ "logps/chosen": -34.49349594116211,
1489
+ "logps/rejected": -30.996051788330078,
1490
+ "loss": 0.6529,
1491
+ "rewards/accuracies": 0.59375,
1492
+ "rewards/chosen": -0.05104157701134682,
1493
+ "rewards/margins": 0.09222711622714996,
1494
+ "rewards/rejected": -0.14326870441436768,
1495
+ "step": 99
1496
+ },
1497
+ {
1498
+ "epoch": 0.9142857142857143,
1499
+ "grad_norm": 1.144099473953247,
1500
+ "learning_rate": 6.187276263508168e-08,
1501
+ "logits/chosen": 0.9783251285552979,
1502
+ "logits/rejected": 0.9752915501594543,
1503
+ "logps/chosen": -24.905807495117188,
1504
+ "logps/rejected": -37.7324104309082,
1505
+ "loss": 0.6312,
1506
+ "rewards/accuracies": 0.640625,
1507
+ "rewards/chosen": -0.022566627711057663,
1508
+ "rewards/margins": 0.14175723493099213,
1509
+ "rewards/rejected": -0.1643238663673401,
1510
+ "step": 100
1511
+ },
1512
+ {
1513
+ "epoch": 0.9234285714285714,
1514
+ "grad_norm": 1.1088513135910034,
1515
+ "learning_rate": 5.018298189593368e-08,
1516
+ "logits/chosen": 0.905106246471405,
1517
+ "logits/rejected": 0.8990155458450317,
1518
+ "logps/chosen": -25.45615577697754,
1519
+ "logps/rejected": -30.45146942138672,
1520
+ "loss": 0.6443,
1521
+ "rewards/accuracies": 0.640625,
1522
+ "rewards/chosen": -0.036811549216508865,
1523
+ "rewards/margins": 0.10829822719097137,
1524
+ "rewards/rejected": -0.14510978758335114,
1525
+ "step": 101
1526
+ },
1527
+ {
1528
+ "epoch": 0.9325714285714286,
1529
+ "grad_norm": 1.3354872465133667,
1530
+ "learning_rate": 3.969748782418991e-08,
1531
+ "logits/chosen": 0.9115935564041138,
1532
+ "logits/rejected": 0.9136921167373657,
1533
+ "logps/chosen": -27.525684356689453,
1534
+ "logps/rejected": -37.691627502441406,
1535
+ "loss": 0.6263,
1536
+ "rewards/accuracies": 0.734375,
1537
+ "rewards/chosen": -0.05684167891740799,
1538
+ "rewards/margins": 0.14856788516044617,
1539
+ "rewards/rejected": -0.20540957152843475,
1540
+ "step": 102
1541
+ },
1542
+ {
1543
+ "epoch": 0.9417142857142857,
1544
+ "grad_norm": 1.065632700920105,
1545
+ "learning_rate": 3.042499016773881e-08,
1546
+ "logits/chosen": 0.967872142791748,
1547
+ "logits/rejected": 0.9852058291435242,
1548
+ "logps/chosen": -33.24607849121094,
1549
+ "logps/rejected": -37.51969528198242,
1550
+ "loss": 0.6414,
1551
+ "rewards/accuracies": 0.65625,
1552
+ "rewards/chosen": -0.015309065580368042,
1553
+ "rewards/margins": 0.1178901419043541,
1554
+ "rewards/rejected": -0.13319920003414154,
1555
+ "step": 103
1556
+ },
1557
+ {
1558
+ "epoch": 0.9508571428571428,
1559
+ "grad_norm": 1.596121907234192,
1560
+ "learning_rate": 2.2373191102207647e-08,
1561
+ "logits/chosen": 0.8452604413032532,
1562
+ "logits/rejected": 0.8531680107116699,
1563
+ "logps/chosen": -51.22788619995117,
1564
+ "logps/rejected": -39.669349670410156,
1565
+ "loss": 0.6496,
1566
+ "rewards/accuracies": 0.53125,
1567
+ "rewards/chosen": -0.04967741668224335,
1568
+ "rewards/margins": 0.0980379655957222,
1569
+ "rewards/rejected": -0.14771538972854614,
1570
+ "step": 104
1571
+ },
1572
+ {
1573
+ "epoch": 0.96,
1574
+ "grad_norm": 1.0667520761489868,
1575
+ "learning_rate": 1.5548778833171463e-08,
1576
+ "logits/chosen": 0.8063293695449829,
1577
+ "logits/rejected": 0.8101313710212708,
1578
+ "logps/chosen": -20.789138793945312,
1579
+ "logps/rejected": -28.230112075805664,
1580
+ "loss": 0.6416,
1581
+ "rewards/accuracies": 0.59375,
1582
+ "rewards/chosen": -0.04012560844421387,
1583
+ "rewards/margins": 0.11673933267593384,
1584
+ "rewards/rejected": -0.1568649411201477,
1585
+ "step": 105
1586
+ },
1587
+ {
1588
+ "epoch": 0.9691428571428572,
1589
+ "grad_norm": 1.2553832530975342,
1590
+ "learning_rate": 9.957422040612507e-09,
1591
+ "logits/chosen": 0.857400119304657,
1592
+ "logits/rejected": 0.8552380204200745,
1593
+ "logps/chosen": -30.64198112487793,
1594
+ "logps/rejected": -33.72669982910156,
1595
+ "loss": 0.6445,
1596
+ "rewards/accuracies": 0.625,
1597
+ "rewards/chosen": -0.020448889583349228,
1598
+ "rewards/margins": 0.1077820211648941,
1599
+ "rewards/rejected": -0.12823091447353363,
1600
+ "step": 106
1601
+ },
1602
+ {
1603
+ "epoch": 0.9782857142857143,
1604
+ "grad_norm": 1.1201170682907104,
1605
+ "learning_rate": 5.6037651702463e-09,
1606
+ "logits/chosen": 0.9444741010665894,
1607
+ "logits/rejected": 0.9209635257720947,
1608
+ "logps/chosen": -29.356178283691406,
1609
+ "logps/rejected": -32.059410095214844,
1610
+ "loss": 0.6312,
1611
+ "rewards/accuracies": 0.765625,
1612
+ "rewards/chosen": -0.027592379599809647,
1613
+ "rewards/margins": 0.13553185760974884,
1614
+ "rewards/rejected": -0.16312424838542938,
1615
+ "step": 107
1616
+ },
1617
+ {
1618
+ "epoch": 0.9874285714285714,
1619
+ "grad_norm": 1.1855032444000244,
1620
+ "learning_rate": 2.491424575625123e-09,
1621
+ "logits/chosen": 0.9110236763954163,
1622
+ "logits/rejected": 0.923471212387085,
1623
+ "logps/chosen": -27.340978622436523,
1624
+ "logps/rejected": -34.48983383178711,
1625
+ "loss": 0.6424,
1626
+ "rewards/accuracies": 0.65625,
1627
+ "rewards/chosen": -0.04438486695289612,
1628
+ "rewards/margins": 0.11192703992128372,
1629
+ "rewards/rejected": -0.15631189942359924,
1630
+ "step": 108
1631
+ },
1632
+ {
1633
+ "epoch": 0.9965714285714286,
1634
+ "grad_norm": 1.2550923824310303,
1635
+ "learning_rate": 6.229855142232399e-10,
1636
+ "logits/chosen": 0.9552398324012756,
1637
+ "logits/rejected": 0.965080976486206,
1638
+ "logps/chosen": -34.74972915649414,
1639
+ "logps/rejected": -46.57404327392578,
1640
+ "loss": 0.6358,
1641
+ "rewards/accuracies": 0.640625,
1642
+ "rewards/chosen": -0.03221715986728668,
1643
+ "rewards/margins": 0.13598641753196716,
1644
+ "rewards/rejected": -0.16820359230041504,
1645
+ "step": 109
1646
+ },
1647
+ {
1648
+ "epoch": 0.9965714285714286,
1649
+ "step": 109,
1650
+ "total_flos": 4.2490128934476186e+17,
1651
+ "train_loss": 0.666442971710765,
1652
+ "train_runtime": 4023.4671,
1653
+ "train_samples_per_second": 1.74,
1654
+ "train_steps_per_second": 0.027
1655
+ }
1656
+ ],
1657
+ "logging_steps": 1.0,
1658
+ "max_steps": 109,
1659
+ "num_input_tokens_seen": 0,
1660
+ "num_train_epochs": 1,
1661
+ "save_steps": 1000,
1662
+ "stateful_callbacks": {
1663
+ "TrainerControl": {
1664
+ "args": {
1665
+ "should_epoch_stop": false,
1666
+ "should_evaluate": false,
1667
+ "should_log": false,
1668
+ "should_save": true,
1669
+ "should_training_stop": true
1670
+ },
1671
+ "attributes": {}
1672
+ }
1673
+ },
1674
+ "total_flos": 4.2490128934476186e+17,
1675
+ "train_batch_size": 1,
1676
+ "trial_name": null,
1677
+ "trial_params": null
1678
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c24b1a1fb7f8f46b9605639ada91f99b5db38bd03dc52d2b80b2ffd79e387952
3
+ size 7480
vocab.json ADDED
The diff for this file is too large to render. See raw diff