BraylonDash commited on
Commit
7324230
1 Parent(s): 939440e

Model save

Browse files
README.md ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ tags:
4
+ - trl
5
+ - dpo
6
+ - generated_from_trainer
7
+ base_model: DUAL-GPO/phi-2-ipo-chatml-lora-20k-40k-i1-merged
8
+ model-index:
9
+ - name: phi-2-ipo-chatml-lora-40k-60k-i2
10
+ results: []
11
+ ---
12
+
13
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
+ should probably proofread and complete it, then remove this comment. -->
15
+
16
+ # phi-2-ipo-chatml-lora-40k-60k-i2
17
+
18
+ This model is a fine-tuned version of [DUAL-GPO/phi-2-ipo-chatml-lora-20k-40k-i1-merged](https://huggingface.co/DUAL-GPO/phi-2-ipo-chatml-lora-20k-40k-i1-merged) on the None dataset.
19
+
20
+ ## Model description
21
+
22
+ More information needed
23
+
24
+ ## Intended uses & limitations
25
+
26
+ More information needed
27
+
28
+ ## Training and evaluation data
29
+
30
+ More information needed
31
+
32
+ ## Training procedure
33
+
34
+ ### Training hyperparameters
35
+
36
+ The following hyperparameters were used during training:
37
+ - learning_rate: 5e-06
38
+ - train_batch_size: 4
39
+ - eval_batch_size: 4
40
+ - seed: 42
41
+ - distributed_type: multi-GPU
42
+ - gradient_accumulation_steps: 4
43
+ - total_train_batch_size: 16
44
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
+ - lr_scheduler_type: cosine
46
+ - lr_scheduler_warmup_ratio: 0.1
47
+ - num_epochs: 1
48
+
49
+ ### Training results
50
+
51
+
52
+
53
+ ### Framework versions
54
+
55
+ - PEFT 0.7.1
56
+ - Transformers 4.36.2
57
+ - Pytorch 2.1.2
58
+ - Datasets 2.14.6
59
+ - Tokenizers 0.15.2
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c314235a83640c2dc0f66481534146937295ccb386f25e4be28b6e8b3c44db8c
3
  size 335579632
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6caddbe9e2f9059555dd64519e17815652b41e3d58e1c584f7e228a89ab5f31a
3
  size 335579632
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "train_loss": 219325.9635,
4
+ "train_runtime": 11890.0433,
5
+ "train_samples": 20000,
6
+ "train_samples_per_second": 1.682,
7
+ "train_steps_per_second": 0.105
8
+ }
emissions.csv ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ timestamp,project_name,run_id,duration,emissions,emissions_rate,cpu_power,gpu_power,ram_power,cpu_energy,gpu_energy,ram_energy,energy_consumed,country_name,country_iso_code,region,cloud_provider,cloud_region,os,python_version,codecarbon_version,cpu_count,cpu_model,gpu_count,gpu_model,longitude,latitude,ram_total_size,tracking_mode,on_cloud,pue
2
+ 2024-09-18T02:21:52,codecarbon,516cb654-2390-4b86-9641-ec3dbf099fa9,11890.04763507843,1.8539184266675317,0.00015592186705779356,42.5,452.443,188.74309015274048,0.14036677627960856,2.7278591478401344,0.6215225858342562,3.4897485099539987,Australia,AUS,new south wales,,,Linux-5.15.0-84-generic-x86_64-with-glibc2.35,3.10.14,2.2.3,32,Intel(R) Xeon(R) W-3335 CPU @ 3.40GHz,4,4 x NVIDIA GeForce RTX 4090,151.1919,-33.887,503.3149070739746,machine,N,1.0
runs/Sep17_23-03-02_gpu4-119-5/events.out.tfevents.1726578222.gpu4-119-5.2890131.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0d14d96e5efd41dee9bec833b87ecbb0aeb804b9d19f4bd77a55edd6e0a93342
3
- size 81533
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a0520d8775cef0d56e133529850662377f4dc8b3d88b4a121fe01dd25765a7a3
3
+ size 85057
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "train_loss": 219325.9635,
4
+ "train_runtime": 11890.0433,
5
+ "train_samples": 20000,
6
+ "train_samples_per_second": 1.682,
7
+ "train_steps_per_second": 0.105
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1794 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 1250,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0,
13
+ "learning_rate": 4e-08,
14
+ "logits/chosen": 0.20467731356620789,
15
+ "logits/rejected": -0.03319101780653,
16
+ "logps/chosen": -809.8218383789062,
17
+ "logps/rejected": -799.7120361328125,
18
+ "loss": 250000.0,
19
+ "rewards/accuracies": 0.0,
20
+ "rewards/chosen": 0.0,
21
+ "rewards/margins": 0.0,
22
+ "rewards/rejected": 0.0,
23
+ "step": 1
24
+ },
25
+ {
26
+ "epoch": 0.01,
27
+ "learning_rate": 4.0000000000000003e-07,
28
+ "logits/chosen": 0.13138051331043243,
29
+ "logits/rejected": 0.1720053106546402,
30
+ "logps/chosen": -657.9158325195312,
31
+ "logps/rejected": -722.3764038085938,
32
+ "loss": 250003.4722,
33
+ "rewards/accuracies": 0.3680555522441864,
34
+ "rewards/chosen": -5.69360563531518e-05,
35
+ "rewards/margins": -4.6177754597920284e-07,
36
+ "rewards/rejected": -5.6474262237316e-05,
37
+ "step": 10
38
+ },
39
+ {
40
+ "epoch": 0.02,
41
+ "learning_rate": 8.000000000000001e-07,
42
+ "logits/chosen": 0.08665986359119415,
43
+ "logits/rejected": 0.11624719947576523,
44
+ "logps/chosen": -698.6005249023438,
45
+ "logps/rejected": -714.4202880859375,
46
+ "loss": 250123.75,
47
+ "rewards/accuracies": 0.4000000059604645,
48
+ "rewards/chosen": -0.000744153221603483,
49
+ "rewards/margins": -0.00012029742356389761,
50
+ "rewards/rejected": -0.0006238558562472463,
51
+ "step": 20
52
+ },
53
+ {
54
+ "epoch": 0.02,
55
+ "learning_rate": 1.2000000000000002e-06,
56
+ "logits/chosen": 0.09405649453401566,
57
+ "logits/rejected": 0.15623515844345093,
58
+ "logps/chosen": -684.2938842773438,
59
+ "logps/rejected": -714.6024169921875,
60
+ "loss": 249469.95,
61
+ "rewards/accuracies": 0.518750011920929,
62
+ "rewards/chosen": -0.0029351129196584225,
63
+ "rewards/margins": 0.0005373213207349181,
64
+ "rewards/rejected": -0.003472433891147375,
65
+ "step": 30
66
+ },
67
+ {
68
+ "epoch": 0.03,
69
+ "learning_rate": 1.6000000000000001e-06,
70
+ "logits/chosen": 0.05473437160253525,
71
+ "logits/rejected": 0.15249383449554443,
72
+ "logps/chosen": -747.6649780273438,
73
+ "logps/rejected": -777.7386474609375,
74
+ "loss": 248992.375,
75
+ "rewards/accuracies": 0.5,
76
+ "rewards/chosen": -0.0077500538900494576,
77
+ "rewards/margins": 0.0010270824423059821,
78
+ "rewards/rejected": -0.008777135983109474,
79
+ "step": 40
80
+ },
81
+ {
82
+ "epoch": 0.04,
83
+ "learning_rate": 2.0000000000000003e-06,
84
+ "logits/chosen": 0.16803810000419617,
85
+ "logits/rejected": 0.15039989352226257,
86
+ "logps/chosen": -727.0869750976562,
87
+ "logps/rejected": -752.9498291015625,
88
+ "loss": 247123.8,
89
+ "rewards/accuracies": 0.581250011920929,
90
+ "rewards/chosen": -0.01702238991856575,
91
+ "rewards/margins": 0.0029481148812919855,
92
+ "rewards/rejected": -0.019970500841736794,
93
+ "step": 50
94
+ },
95
+ {
96
+ "epoch": 0.05,
97
+ "learning_rate": 2.4000000000000003e-06,
98
+ "logits/chosen": 0.061672769486904144,
99
+ "logits/rejected": 0.12831617891788483,
100
+ "logps/chosen": -744.3715209960938,
101
+ "logps/rejected": -771.6224365234375,
102
+ "loss": 245689.85,
103
+ "rewards/accuracies": 0.550000011920929,
104
+ "rewards/chosen": -0.03865597024559975,
105
+ "rewards/margins": 0.004632375668734312,
106
+ "rewards/rejected": -0.04328834265470505,
107
+ "step": 60
108
+ },
109
+ {
110
+ "epoch": 0.06,
111
+ "learning_rate": 2.8000000000000003e-06,
112
+ "logits/chosen": 0.11488649994134903,
113
+ "logits/rejected": 0.17836399376392365,
114
+ "logps/chosen": -778.160400390625,
115
+ "logps/rejected": -824.7205200195312,
116
+ "loss": 241170.575,
117
+ "rewards/accuracies": 0.543749988079071,
118
+ "rewards/chosen": -0.06250237673521042,
119
+ "rewards/margins": 0.009986966848373413,
120
+ "rewards/rejected": -0.07248934358358383,
121
+ "step": 70
122
+ },
123
+ {
124
+ "epoch": 0.06,
125
+ "learning_rate": 3.2000000000000003e-06,
126
+ "logits/chosen": 0.10301940143108368,
127
+ "logits/rejected": 0.1564978063106537,
128
+ "logps/chosen": -789.8165283203125,
129
+ "logps/rejected": -818.8817138671875,
130
+ "loss": 242407.825,
131
+ "rewards/accuracies": 0.5625,
132
+ "rewards/chosen": -0.1025240570306778,
133
+ "rewards/margins": 0.010024911724030972,
134
+ "rewards/rejected": -0.11254896223545074,
135
+ "step": 80
136
+ },
137
+ {
138
+ "epoch": 0.07,
139
+ "learning_rate": 3.6000000000000003e-06,
140
+ "logits/chosen": 0.10446761548519135,
141
+ "logits/rejected": 0.13645999133586884,
142
+ "logps/chosen": -806.6678466796875,
143
+ "logps/rejected": -859.4196166992188,
144
+ "loss": 247294.85,
145
+ "rewards/accuracies": 0.46875,
146
+ "rewards/chosen": -0.10863246023654938,
147
+ "rewards/margins": 0.005576482508331537,
148
+ "rewards/rejected": -0.11420895159244537,
149
+ "step": 90
150
+ },
151
+ {
152
+ "epoch": 0.08,
153
+ "learning_rate": 4.000000000000001e-06,
154
+ "logits/chosen": 0.08340464532375336,
155
+ "logits/rejected": 0.09231794625520706,
156
+ "logps/chosen": -788.7113037109375,
157
+ "logps/rejected": -876.8160400390625,
158
+ "loss": 232655.8,
159
+ "rewards/accuracies": 0.5874999761581421,
160
+ "rewards/chosen": -0.08170753717422485,
161
+ "rewards/margins": 0.019764257594943047,
162
+ "rewards/rejected": -0.10147178173065186,
163
+ "step": 100
164
+ },
165
+ {
166
+ "epoch": 0.09,
167
+ "learning_rate": 4.4e-06,
168
+ "logits/chosen": 0.07701124250888824,
169
+ "logits/rejected": 0.1278785914182663,
170
+ "logps/chosen": -759.1944580078125,
171
+ "logps/rejected": -814.4751586914062,
172
+ "loss": 240107.05,
173
+ "rewards/accuracies": 0.53125,
174
+ "rewards/chosen": -0.07757158577442169,
175
+ "rewards/margins": 0.013457924127578735,
176
+ "rewards/rejected": -0.09102950990200043,
177
+ "step": 110
178
+ },
179
+ {
180
+ "epoch": 0.1,
181
+ "learning_rate": 4.800000000000001e-06,
182
+ "logits/chosen": 0.06585706770420074,
183
+ "logits/rejected": 0.16044208407402039,
184
+ "logps/chosen": -753.0533447265625,
185
+ "logps/rejected": -833.3118896484375,
186
+ "loss": 238825.4,
187
+ "rewards/accuracies": 0.53125,
188
+ "rewards/chosen": -0.07277105003595352,
189
+ "rewards/margins": 0.014289605431258678,
190
+ "rewards/rejected": -0.08706065267324448,
191
+ "step": 120
192
+ },
193
+ {
194
+ "epoch": 0.1,
195
+ "learning_rate": 4.999756310023261e-06,
196
+ "logits/chosen": 0.14829027652740479,
197
+ "logits/rejected": 0.173395037651062,
198
+ "logps/chosen": -817.2315673828125,
199
+ "logps/rejected": -921.9486083984375,
200
+ "loss": 229325.1,
201
+ "rewards/accuracies": 0.625,
202
+ "rewards/chosen": -0.10191284120082855,
203
+ "rewards/margins": 0.0245502982288599,
204
+ "rewards/rejected": -0.1264631450176239,
205
+ "step": 130
206
+ },
207
+ {
208
+ "epoch": 0.11,
209
+ "learning_rate": 4.997807075247147e-06,
210
+ "logits/chosen": 0.1464642435312271,
211
+ "logits/rejected": 0.15901242196559906,
212
+ "logps/chosen": -863.0106201171875,
213
+ "logps/rejected": -923.87353515625,
214
+ "loss": 225025.15,
215
+ "rewards/accuracies": 0.6187499761581421,
216
+ "rewards/chosen": -0.12692296504974365,
217
+ "rewards/margins": 0.03183966130018234,
218
+ "rewards/rejected": -0.1587626188993454,
219
+ "step": 140
220
+ },
221
+ {
222
+ "epoch": 0.12,
223
+ "learning_rate": 4.993910125649561e-06,
224
+ "logits/chosen": 0.20590683817863464,
225
+ "logits/rejected": 0.2767329812049866,
226
+ "logps/chosen": -809.0446166992188,
227
+ "logps/rejected": -827.0875854492188,
228
+ "loss": 247641.325,
229
+ "rewards/accuracies": 0.53125,
230
+ "rewards/chosen": -0.14790478348731995,
231
+ "rewards/margins": 0.013205945491790771,
232
+ "rewards/rejected": -0.1611107438802719,
233
+ "step": 150
234
+ },
235
+ {
236
+ "epoch": 0.13,
237
+ "learning_rate": 4.988068499954578e-06,
238
+ "logits/chosen": 0.1629045307636261,
239
+ "logits/rejected": 0.21712151169776917,
240
+ "logps/chosen": -841.6968994140625,
241
+ "logps/rejected": -926.6705322265625,
242
+ "loss": 228410.175,
243
+ "rewards/accuracies": 0.543749988079071,
244
+ "rewards/chosen": -0.13649746775627136,
245
+ "rewards/margins": 0.029097670689225197,
246
+ "rewards/rejected": -0.1655951291322708,
247
+ "step": 160
248
+ },
249
+ {
250
+ "epoch": 0.14,
251
+ "learning_rate": 4.980286753286196e-06,
252
+ "logits/chosen": 0.1644325703382492,
253
+ "logits/rejected": 0.19548630714416504,
254
+ "logps/chosen": -936.2694091796875,
255
+ "logps/rejected": -998.8902587890625,
256
+ "loss": 239878.925,
257
+ "rewards/accuracies": 0.5625,
258
+ "rewards/chosen": -0.22792987525463104,
259
+ "rewards/margins": 0.02233467996120453,
260
+ "rewards/rejected": -0.25026455521583557,
261
+ "step": 170
262
+ },
263
+ {
264
+ "epoch": 0.14,
265
+ "learning_rate": 4.970570953616383e-06,
266
+ "logits/chosen": 0.19817355275154114,
267
+ "logits/rejected": 0.1892698109149933,
268
+ "logps/chosen": -947.0306396484375,
269
+ "logps/rejected": -1044.988037109375,
270
+ "loss": 224134.05,
271
+ "rewards/accuracies": 0.5874999761581421,
272
+ "rewards/chosen": -0.24009647965431213,
273
+ "rewards/margins": 0.04032464697957039,
274
+ "rewards/rejected": -0.28042110800743103,
275
+ "step": 180
276
+ },
277
+ {
278
+ "epoch": 0.15,
279
+ "learning_rate": 4.958928677033465e-06,
280
+ "logits/chosen": 0.2510104477405548,
281
+ "logits/rejected": 0.2124006450176239,
282
+ "logps/chosen": -844.8209838867188,
283
+ "logps/rejected": -940.4304809570312,
284
+ "loss": 227132.5,
285
+ "rewards/accuracies": 0.5687500238418579,
286
+ "rewards/chosen": -0.15492157638072968,
287
+ "rewards/margins": 0.03421669453382492,
288
+ "rewards/rejected": -0.189138263463974,
289
+ "step": 190
290
+ },
291
+ {
292
+ "epoch": 0.16,
293
+ "learning_rate": 4.9453690018345144e-06,
294
+ "logits/chosen": 0.2112915813922882,
295
+ "logits/rejected": 0.26694074273109436,
296
+ "logps/chosen": -857.0277099609375,
297
+ "logps/rejected": -930.4342041015625,
298
+ "loss": 226498.725,
299
+ "rewards/accuracies": 0.612500011920929,
300
+ "rewards/chosen": -0.15992014110088348,
301
+ "rewards/margins": 0.036206770688295364,
302
+ "rewards/rejected": -0.19612692296504974,
303
+ "step": 200
304
+ },
305
+ {
306
+ "epoch": 0.17,
307
+ "learning_rate": 4.9299025014463665e-06,
308
+ "logits/chosen": 0.18406227231025696,
309
+ "logits/rejected": 0.323600709438324,
310
+ "logps/chosen": -980.2025146484375,
311
+ "logps/rejected": -1063.402099609375,
312
+ "loss": 226366.0,
313
+ "rewards/accuracies": 0.5874999761581421,
314
+ "rewards/chosen": -0.2587742209434509,
315
+ "rewards/margins": 0.04289933666586876,
316
+ "rewards/rejected": -0.3016735911369324,
317
+ "step": 210
318
+ },
319
+ {
320
+ "epoch": 0.18,
321
+ "learning_rate": 4.912541236180779e-06,
322
+ "logits/chosen": 0.298921138048172,
323
+ "logits/rejected": 0.28891488909721375,
324
+ "logps/chosen": -995.7527465820312,
325
+ "logps/rejected": -1056.0579833984375,
326
+ "loss": 225886.0,
327
+ "rewards/accuracies": 0.5562499761581421,
328
+ "rewards/chosen": -0.27992406487464905,
329
+ "rewards/margins": 0.04216141253709793,
330
+ "rewards/rejected": -0.32208549976348877,
331
+ "step": 220
332
+ },
333
+ {
334
+ "epoch": 0.18,
335
+ "learning_rate": 4.893298743830168e-06,
336
+ "logits/chosen": 0.2236076295375824,
337
+ "logits/rejected": 0.26189810037612915,
338
+ "logps/chosen": -940.28369140625,
339
+ "logps/rejected": -1038.412109375,
340
+ "loss": 221704.9,
341
+ "rewards/accuracies": 0.612500011920929,
342
+ "rewards/chosen": -0.19546031951904297,
343
+ "rewards/margins": 0.0415831059217453,
344
+ "rewards/rejected": -0.23704342544078827,
345
+ "step": 230
346
+ },
347
+ {
348
+ "epoch": 0.19,
349
+ "learning_rate": 4.8721900291112415e-06,
350
+ "logits/chosen": 0.21125178039073944,
351
+ "logits/rejected": 0.2499365508556366,
352
+ "logps/chosen": -828.0550537109375,
353
+ "logps/rejected": -929.4462890625,
354
+ "loss": 224655.025,
355
+ "rewards/accuracies": 0.543749988079071,
356
+ "rewards/chosen": -0.1438567340373993,
357
+ "rewards/margins": 0.03509406000375748,
358
+ "rewards/rejected": -0.17895078659057617,
359
+ "step": 240
360
+ },
361
+ {
362
+ "epoch": 0.2,
363
+ "learning_rate": 4.849231551964771e-06,
364
+ "logits/chosen": 0.2047933042049408,
365
+ "logits/rejected": 0.2937256693840027,
366
+ "logps/chosen": -932.6740112304688,
367
+ "logps/rejected": -978.4197998046875,
368
+ "loss": 231297.9,
369
+ "rewards/accuracies": 0.5062500238418579,
370
+ "rewards/chosen": -0.19101986289024353,
371
+ "rewards/margins": 0.029795492067933083,
372
+ "rewards/rejected": -0.22081534564495087,
373
+ "step": 250
374
+ },
375
+ {
376
+ "epoch": 0.21,
377
+ "learning_rate": 4.824441214720629e-06,
378
+ "logits/chosen": 0.2713222801685333,
379
+ "logits/rejected": 0.32194697856903076,
380
+ "logps/chosen": -980.1368408203125,
381
+ "logps/rejected": -1050.428955078125,
382
+ "loss": 230157.3,
383
+ "rewards/accuracies": 0.606249988079071,
384
+ "rewards/chosen": -0.2551424205303192,
385
+ "rewards/margins": 0.036989547312259674,
386
+ "rewards/rejected": -0.2921319603919983,
387
+ "step": 260
388
+ },
389
+ {
390
+ "epoch": 0.22,
391
+ "learning_rate": 4.7978383481380865e-06,
392
+ "logits/chosen": 0.2948392331600189,
393
+ "logits/rejected": 0.3368261754512787,
394
+ "logps/chosen": -969.5646362304688,
395
+ "logps/rejected": -1044.114990234375,
396
+ "loss": 221846.5,
397
+ "rewards/accuracies": 0.581250011920929,
398
+ "rewards/chosen": -0.2587720453739166,
399
+ "rewards/margins": 0.04627389460802078,
400
+ "rewards/rejected": -0.3050459325313568,
401
+ "step": 270
402
+ },
403
+ {
404
+ "epoch": 0.22,
405
+ "learning_rate": 4.769443696332272e-06,
406
+ "logits/chosen": 0.25841373205184937,
407
+ "logits/rejected": 0.28298312425613403,
408
+ "logps/chosen": -966.0433349609375,
409
+ "logps/rejected": -1030.305419921875,
410
+ "loss": 225948.625,
411
+ "rewards/accuracies": 0.5625,
412
+ "rewards/chosen": -0.2648239731788635,
413
+ "rewards/margins": 0.037199851125478745,
414
+ "rewards/rejected": -0.30202382802963257,
415
+ "step": 280
416
+ },
417
+ {
418
+ "epoch": 0.23,
419
+ "learning_rate": 4.7392794005985324e-06,
420
+ "logits/chosen": 0.3133973181247711,
421
+ "logits/rejected": 0.3274853229522705,
422
+ "logps/chosen": -996.1575317382812,
423
+ "logps/rejected": -1076.3135986328125,
424
+ "loss": 232305.25,
425
+ "rewards/accuracies": 0.5249999761581421,
426
+ "rewards/chosen": -0.31282705068588257,
427
+ "rewards/margins": 0.04026792198419571,
428
+ "rewards/rejected": -0.35309499502182007,
429
+ "step": 290
430
+ },
431
+ {
432
+ "epoch": 0.24,
433
+ "learning_rate": 4.707368982147318e-06,
434
+ "logits/chosen": 0.2312435805797577,
435
+ "logits/rejected": 0.3617837131023407,
436
+ "logps/chosen": -970.8505859375,
437
+ "logps/rejected": -1032.5286865234375,
438
+ "loss": 218085.225,
439
+ "rewards/accuracies": 0.59375,
440
+ "rewards/chosen": -0.2635905146598816,
441
+ "rewards/margins": 0.047272901982069016,
442
+ "rewards/rejected": -0.3108634054660797,
443
+ "step": 300
444
+ },
445
+ {
446
+ "epoch": 0.25,
447
+ "learning_rate": 4.673737323763048e-06,
448
+ "logits/chosen": 0.2505849003791809,
449
+ "logits/rejected": 0.37897807359695435,
450
+ "logps/chosen": -891.2982177734375,
451
+ "logps/rejected": -984.837890625,
452
+ "loss": 221958.6,
453
+ "rewards/accuracies": 0.5562499761581421,
454
+ "rewards/chosen": -0.21344509720802307,
455
+ "rewards/margins": 0.03914090245962143,
456
+ "rewards/rejected": -0.2525860071182251,
457
+ "step": 310
458
+ },
459
+ {
460
+ "epoch": 0.26,
461
+ "learning_rate": 4.638410650401267e-06,
462
+ "logits/chosen": 0.32055309414863586,
463
+ "logits/rejected": 0.4090685248374939,
464
+ "logps/chosen": -974.4312744140625,
465
+ "logps/rejected": -1051.5267333984375,
466
+ "loss": 220473.5,
467
+ "rewards/accuracies": 0.5562499761581421,
468
+ "rewards/chosen": -0.2853444218635559,
469
+ "rewards/margins": 0.04749929532408714,
470
+ "rewards/rejected": -0.33284372091293335,
471
+ "step": 320
472
+ },
473
+ {
474
+ "epoch": 0.26,
475
+ "learning_rate": 4.601416508739211e-06,
476
+ "logits/chosen": 0.37889987230300903,
477
+ "logits/rejected": 0.3925861716270447,
478
+ "logps/chosen": -1118.357666015625,
479
+ "logps/rejected": -1232.1087646484375,
480
+ "loss": 209207.2,
481
+ "rewards/accuracies": 0.606249988079071,
482
+ "rewards/chosen": -0.3847602605819702,
483
+ "rewards/margins": 0.07533888518810272,
484
+ "rewards/rejected": -0.46009913086891174,
485
+ "step": 330
486
+ },
487
+ {
488
+ "epoch": 0.27,
489
+ "learning_rate": 4.562783745695738e-06,
490
+ "logits/chosen": 0.2892415225505829,
491
+ "logits/rejected": 0.3394598066806793,
492
+ "logps/chosen": -1099.81201171875,
493
+ "logps/rejected": -1189.7840576171875,
494
+ "loss": 217431.375,
495
+ "rewards/accuracies": 0.643750011920929,
496
+ "rewards/chosen": -0.35406026244163513,
497
+ "rewards/margins": 0.05799219757318497,
498
+ "rewards/rejected": -0.4120524525642395,
499
+ "step": 340
500
+ },
501
+ {
502
+ "epoch": 0.28,
503
+ "learning_rate": 4.522542485937369e-06,
504
+ "logits/chosen": 0.27173861861228943,
505
+ "logits/rejected": 0.3863845467567444,
506
+ "logps/chosen": -1021.44140625,
507
+ "logps/rejected": -1087.6248779296875,
508
+ "loss": 228095.375,
509
+ "rewards/accuracies": 0.550000011920929,
510
+ "rewards/chosen": -0.28105270862579346,
511
+ "rewards/margins": 0.03988432139158249,
512
+ "rewards/rejected": -0.32093703746795654,
513
+ "step": 350
514
+ },
515
+ {
516
+ "epoch": 0.29,
517
+ "learning_rate": 4.4807241083879774e-06,
518
+ "logits/chosen": 0.3540768027305603,
519
+ "logits/rejected": 0.37046054005622864,
520
+ "logps/chosen": -963.1799926757812,
521
+ "logps/rejected": -1018.9503784179688,
522
+ "loss": 228921.25,
523
+ "rewards/accuracies": 0.574999988079071,
524
+ "rewards/chosen": -0.25751227140426636,
525
+ "rewards/margins": 0.03808392956852913,
526
+ "rewards/rejected": -0.2955962121486664,
527
+ "step": 360
528
+ },
529
+ {
530
+ "epoch": 0.3,
531
+ "learning_rate": 4.437361221760449e-06,
532
+ "logits/chosen": 0.3426358103752136,
533
+ "logits/rejected": 0.4119463562965393,
534
+ "logps/chosen": -995.5330200195312,
535
+ "logps/rejected": -1035.6353759765625,
536
+ "loss": 226486.05,
537
+ "rewards/accuracies": 0.5625,
538
+ "rewards/chosen": -0.26880544424057007,
539
+ "rewards/margins": 0.03966186195611954,
540
+ "rewards/rejected": -0.3084673285484314,
541
+ "step": 370
542
+ },
543
+ {
544
+ "epoch": 0.3,
545
+ "learning_rate": 4.3924876391293915e-06,
546
+ "logits/chosen": 0.36900123953819275,
547
+ "logits/rejected": 0.35201960802078247,
548
+ "logps/chosen": -936.4334106445312,
549
+ "logps/rejected": -1031.63037109375,
550
+ "loss": 222121.7,
551
+ "rewards/accuracies": 0.581250011920929,
552
+ "rewards/chosen": -0.24716052412986755,
553
+ "rewards/margins": 0.043841078877449036,
554
+ "rewards/rejected": -0.2910016179084778,
555
+ "step": 380
556
+ },
557
+ {
558
+ "epoch": 0.31,
559
+ "learning_rate": 4.346138351564711e-06,
560
+ "logits/chosen": 0.3757195770740509,
561
+ "logits/rejected": 0.4981180727481842,
562
+ "logps/chosen": -1056.494384765625,
563
+ "logps/rejected": -1187.510986328125,
564
+ "loss": 199957.3375,
565
+ "rewards/accuracies": 0.675000011920929,
566
+ "rewards/chosen": -0.30984193086624146,
567
+ "rewards/margins": 0.07408682256937027,
568
+ "rewards/rejected": -0.3839287757873535,
569
+ "step": 390
570
+ },
571
+ {
572
+ "epoch": 0.32,
573
+ "learning_rate": 4.2983495008466285e-06,
574
+ "logits/chosen": 0.40311890840530396,
575
+ "logits/rejected": 0.4448017179965973,
576
+ "logps/chosen": -1068.086181640625,
577
+ "logps/rejected": -1157.872802734375,
578
+ "loss": 228107.1,
579
+ "rewards/accuracies": 0.574999988079071,
580
+ "rewards/chosen": -0.3505800664424896,
581
+ "rewards/margins": 0.054390739649534225,
582
+ "rewards/rejected": -0.40497082471847534,
583
+ "step": 400
584
+ },
585
+ {
586
+ "epoch": 0.33,
587
+ "learning_rate": 4.249158351283414e-06,
588
+ "logits/chosen": 0.37398719787597656,
589
+ "logits/rejected": 0.4472652077674866,
590
+ "logps/chosen": -968.9822387695312,
591
+ "logps/rejected": -1006.4807739257812,
592
+ "loss": 229766.15,
593
+ "rewards/accuracies": 0.550000011920929,
594
+ "rewards/chosen": -0.25533100962638855,
595
+ "rewards/margins": 0.03805527463555336,
596
+ "rewards/rejected": -0.293386310338974,
597
+ "step": 410
598
+ },
599
+ {
600
+ "epoch": 0.34,
601
+ "learning_rate": 4.198603260653792e-06,
602
+ "logits/chosen": 0.3892027735710144,
603
+ "logits/rejected": 0.48470592498779297,
604
+ "logps/chosen": -970.4156494140625,
605
+ "logps/rejected": -1086.793212890625,
606
+ "loss": 213715.4,
607
+ "rewards/accuracies": 0.59375,
608
+ "rewards/chosen": -0.2887799143791199,
609
+ "rewards/margins": 0.05600409582257271,
610
+ "rewards/rejected": -0.3447840213775635,
611
+ "step": 420
612
+ },
613
+ {
614
+ "epoch": 0.34,
615
+ "learning_rate": 4.146723650296701e-06,
616
+ "logits/chosen": 0.3906274437904358,
617
+ "logits/rejected": 0.4788398742675781,
618
+ "logps/chosen": -1038.199462890625,
619
+ "logps/rejected": -1115.3878173828125,
620
+ "loss": 226122.0,
621
+ "rewards/accuracies": 0.550000011920929,
622
+ "rewards/chosen": -0.33887821435928345,
623
+ "rewards/margins": 0.04640539363026619,
624
+ "rewards/rejected": -0.38528358936309814,
625
+ "step": 430
626
+ },
627
+ {
628
+ "epoch": 0.35,
629
+ "learning_rate": 4.093559974371725e-06,
630
+ "logits/chosen": 0.4500131607055664,
631
+ "logits/rejected": 0.4698792099952698,
632
+ "logps/chosen": -999.3037109375,
633
+ "logps/rejected": -1112.6219482421875,
634
+ "loss": 220416.5,
635
+ "rewards/accuracies": 0.5874999761581421,
636
+ "rewards/chosen": -0.32525134086608887,
637
+ "rewards/margins": 0.04823971912264824,
638
+ "rewards/rejected": -0.373491108417511,
639
+ "step": 440
640
+ },
641
+ {
642
+ "epoch": 0.36,
643
+ "learning_rate": 4.039153688314146e-06,
644
+ "logits/chosen": 0.3521023094654083,
645
+ "logits/rejected": 0.4308967590332031,
646
+ "logps/chosen": -1049.7392578125,
647
+ "logps/rejected": -1155.595458984375,
648
+ "loss": 226787.7,
649
+ "rewards/accuracies": 0.518750011920929,
650
+ "rewards/chosen": -0.3899618685245514,
651
+ "rewards/margins": 0.05192166566848755,
652
+ "rewards/rejected": -0.44188356399536133,
653
+ "step": 450
654
+ },
655
+ {
656
+ "epoch": 0.37,
657
+ "learning_rate": 3.983547216509254e-06,
658
+ "logits/chosen": 0.4046708047389984,
659
+ "logits/rejected": 0.44424524903297424,
660
+ "logps/chosen": -1089.29052734375,
661
+ "logps/rejected": -1205.33837890625,
662
+ "loss": 206882.4125,
663
+ "rewards/accuracies": 0.6187499761581421,
664
+ "rewards/chosen": -0.3721011281013489,
665
+ "rewards/margins": 0.06494136154651642,
666
+ "rewards/rejected": -0.4370424747467041,
667
+ "step": 460
668
+ },
669
+ {
670
+ "epoch": 0.38,
671
+ "learning_rate": 3.92678391921108e-06,
672
+ "logits/chosen": 0.43136972188949585,
673
+ "logits/rejected": 0.538188636302948,
674
+ "logps/chosen": -1152.017822265625,
675
+ "logps/rejected": -1202.035888671875,
676
+ "loss": 229295.15,
677
+ "rewards/accuracies": 0.5562499761581421,
678
+ "rewards/chosen": -0.40614837408065796,
679
+ "rewards/margins": 0.04618660733103752,
680
+ "rewards/rejected": -0.45233502984046936,
681
+ "step": 470
682
+ },
683
+ {
684
+ "epoch": 0.38,
685
+ "learning_rate": 3.868908058731376e-06,
686
+ "logits/chosen": 0.5136951804161072,
687
+ "logits/rejected": 0.5672486424446106,
688
+ "logps/chosen": -1073.908203125,
689
+ "logps/rejected": -1192.328125,
690
+ "loss": 214087.0,
691
+ "rewards/accuracies": 0.53125,
692
+ "rewards/chosen": -0.39078179001808167,
693
+ "rewards/margins": 0.06803762912750244,
694
+ "rewards/rejected": -0.4588193893432617,
695
+ "step": 480
696
+ },
697
+ {
698
+ "epoch": 0.39,
699
+ "learning_rate": 3.8099647649251984e-06,
700
+ "logits/chosen": 0.45567411184310913,
701
+ "logits/rejected": 0.5252825021743774,
702
+ "logps/chosen": -1133.7230224609375,
703
+ "logps/rejected": -1230.4957275390625,
704
+ "loss": 224336.625,
705
+ "rewards/accuracies": 0.581250011920929,
706
+ "rewards/chosen": -0.42907077074050903,
707
+ "rewards/margins": 0.056711066514253616,
708
+ "rewards/rejected": -0.48578181862831116,
709
+ "step": 490
710
+ },
711
+ {
712
+ "epoch": 0.4,
713
+ "learning_rate": 3.7500000000000005e-06,
714
+ "logits/chosen": 0.45971164107322693,
715
+ "logits/rejected": 0.42342036962509155,
716
+ "logps/chosen": -1157.3365478515625,
717
+ "logps/rejected": -1285.06982421875,
718
+ "loss": 208970.9,
719
+ "rewards/accuracies": 0.6312500238418579,
720
+ "rewards/chosen": -0.41447123885154724,
721
+ "rewards/margins": 0.07142941653728485,
722
+ "rewards/rejected": -0.48590072989463806,
723
+ "step": 500
724
+ },
725
+ {
726
+ "epoch": 0.41,
727
+ "learning_rate": 3.689060522675689e-06,
728
+ "logits/chosen": 0.4447771906852722,
729
+ "logits/rejected": 0.5252470970153809,
730
+ "logps/chosen": -1077.148681640625,
731
+ "logps/rejected": -1190.3822021484375,
732
+ "loss": 209434.125,
733
+ "rewards/accuracies": 0.6187499761581421,
734
+ "rewards/chosen": -0.37661683559417725,
735
+ "rewards/margins": 0.06856775283813477,
736
+ "rewards/rejected": -0.4451845586299896,
737
+ "step": 510
738
+ },
739
+ {
740
+ "epoch": 0.42,
741
+ "learning_rate": 3.627193851723577e-06,
742
+ "logits/chosen": 0.4108400344848633,
743
+ "logits/rejected": 0.5407587885856628,
744
+ "logps/chosen": -1113.6610107421875,
745
+ "logps/rejected": -1170.6397705078125,
746
+ "loss": 232291.675,
747
+ "rewards/accuracies": 0.5,
748
+ "rewards/chosen": -0.38236182928085327,
749
+ "rewards/margins": 0.04210672527551651,
750
+ "rewards/rejected": -0.4244685769081116,
751
+ "step": 520
752
+ },
753
+ {
754
+ "epoch": 0.42,
755
+ "learning_rate": 3.564448228912682e-06,
756
+ "logits/chosen": 0.4384753108024597,
757
+ "logits/rejected": 0.46903926134109497,
758
+ "logps/chosen": -1154.0391845703125,
759
+ "logps/rejected": -1248.3861083984375,
760
+ "loss": 207475.85,
761
+ "rewards/accuracies": 0.581250011920929,
762
+ "rewards/chosen": -0.4249592423439026,
763
+ "rewards/margins": 0.07294295728206635,
764
+ "rewards/rejected": -0.4979022145271301,
765
+ "step": 530
766
+ },
767
+ {
768
+ "epoch": 0.43,
769
+ "learning_rate": 3.5008725813922383e-06,
770
+ "logits/chosen": 0.38299837708473206,
771
+ "logits/rejected": 0.5994570255279541,
772
+ "logps/chosen": -1220.1324462890625,
773
+ "logps/rejected": -1281.889892578125,
774
+ "loss": 223302.4,
775
+ "rewards/accuracies": 0.581250011920929,
776
+ "rewards/chosen": -0.47278809547424316,
777
+ "rewards/margins": 0.05748339742422104,
778
+ "rewards/rejected": -0.5302715301513672,
779
+ "step": 540
780
+ },
781
+ {
782
+ "epoch": 0.44,
783
+ "learning_rate": 3.436516483539781e-06,
784
+ "logits/chosen": 0.41017085313796997,
785
+ "logits/rejected": 0.5024768114089966,
786
+ "logps/chosen": -1120.16015625,
787
+ "logps/rejected": -1186.6806640625,
788
+ "loss": 207700.575,
789
+ "rewards/accuracies": 0.6312500238418579,
790
+ "rewards/chosen": -0.36709490418434143,
791
+ "rewards/margins": 0.06647273153066635,
792
+ "rewards/rejected": -0.433567613363266,
793
+ "step": 550
794
+ },
795
+ {
796
+ "epoch": 0.45,
797
+ "learning_rate": 3.3714301183045382e-06,
798
+ "logits/chosen": 0.5212954878807068,
799
+ "logits/rejected": 0.4878396987915039,
800
+ "logps/chosen": -1117.4761962890625,
801
+ "logps/rejected": -1205.936279296875,
802
+ "loss": 224293.3,
803
+ "rewards/accuracies": 0.625,
804
+ "rewards/chosen": -0.3602500557899475,
805
+ "rewards/margins": 0.04831800237298012,
806
+ "rewards/rejected": -0.40856799483299255,
807
+ "step": 560
808
+ },
809
+ {
810
+ "epoch": 0.46,
811
+ "learning_rate": 3.3056642380762783e-06,
812
+ "logits/chosen": 0.4528779983520508,
813
+ "logits/rejected": 0.5153405666351318,
814
+ "logps/chosen": -1006.6858520507812,
815
+ "logps/rejected": -1131.8328857421875,
816
+ "loss": 218913.225,
817
+ "rewards/accuracies": 0.59375,
818
+ "rewards/chosen": -0.31067246198654175,
819
+ "rewards/margins": 0.05573346093297005,
820
+ "rewards/rejected": -0.3664059042930603,
821
+ "step": 570
822
+ },
823
+ {
824
+ "epoch": 0.46,
825
+ "learning_rate": 3.2392701251101172e-06,
826
+ "logits/chosen": 0.43213313817977905,
827
+ "logits/rejected": 0.5192700624465942,
828
+ "logps/chosen": -959.2723388671875,
829
+ "logps/rejected": -1070.766357421875,
830
+ "loss": 208478.4375,
831
+ "rewards/accuracies": 0.6312500238418579,
832
+ "rewards/chosen": -0.29122763872146606,
833
+ "rewards/margins": 0.05716808885335922,
834
+ "rewards/rejected": -0.34839576482772827,
835
+ "step": 580
836
+ },
837
+ {
838
+ "epoch": 0.47,
839
+ "learning_rate": 3.1722995515381644e-06,
840
+ "logits/chosen": 0.4300867021083832,
841
+ "logits/rejected": 0.4907846450805664,
842
+ "logps/chosen": -1042.2386474609375,
843
+ "logps/rejected": -1146.758056640625,
844
+ "loss": 232486.95,
845
+ "rewards/accuracies": 0.543749988079071,
846
+ "rewards/chosen": -0.3329560160636902,
847
+ "rewards/margins": 0.05041079968214035,
848
+ "rewards/rejected": -0.38336682319641113,
849
+ "step": 590
850
+ },
851
+ {
852
+ "epoch": 0.48,
853
+ "learning_rate": 3.1048047389991693e-06,
854
+ "logits/chosen": 0.3744855523109436,
855
+ "logits/rejected": 0.534927248954773,
856
+ "logps/chosen": -1041.3548583984375,
857
+ "logps/rejected": -1123.1234130859375,
858
+ "loss": 222175.75,
859
+ "rewards/accuracies": 0.581250011920929,
860
+ "rewards/chosen": -0.3181845247745514,
861
+ "rewards/margins": 0.045694101601839066,
862
+ "rewards/rejected": -0.36387860774993896,
863
+ "step": 600
864
+ },
865
+ {
866
+ "epoch": 0.49,
867
+ "learning_rate": 3.0368383179176584e-06,
868
+ "logits/chosen": 0.43926602602005005,
869
+ "logits/rejected": 0.46200305223464966,
870
+ "logps/chosen": -987.1368408203125,
871
+ "logps/rejected": -1085.827880859375,
872
+ "loss": 219539.875,
873
+ "rewards/accuracies": 0.606249988079071,
874
+ "rewards/chosen": -0.30275067687034607,
875
+ "rewards/margins": 0.04839698225259781,
876
+ "rewards/rejected": -0.35114771127700806,
877
+ "step": 610
878
+ },
879
+ {
880
+ "epoch": 0.5,
881
+ "learning_rate": 2.9684532864643123e-06,
882
+ "logits/chosen": 0.42336663603782654,
883
+ "logits/rejected": 0.44354915618896484,
884
+ "logps/chosen": -973.3244018554688,
885
+ "logps/rejected": -1073.771240234375,
886
+ "loss": 204531.35,
887
+ "rewards/accuracies": 0.59375,
888
+ "rewards/chosen": -0.292645126581192,
889
+ "rewards/margins": 0.06315550208091736,
890
+ "rewards/rejected": -0.3558006286621094,
891
+ "step": 620
892
+ },
893
+ {
894
+ "epoch": 0.5,
895
+ "learning_rate": 2.8997029692295875e-06,
896
+ "logits/chosen": 0.3760843276977539,
897
+ "logits/rejected": 0.5375649333000183,
898
+ "logps/chosen": -1022.2794799804688,
899
+ "logps/rejected": -1099.590576171875,
900
+ "loss": 218063.675,
901
+ "rewards/accuracies": 0.581250011920929,
902
+ "rewards/chosen": -0.2922189235687256,
903
+ "rewards/margins": 0.054534755647182465,
904
+ "rewards/rejected": -0.34675365686416626,
905
+ "step": 630
906
+ },
907
+ {
908
+ "epoch": 0.51,
909
+ "learning_rate": 2.8306409756428067e-06,
910
+ "logits/chosen": 0.407614529132843,
911
+ "logits/rejected": 0.4755696654319763,
912
+ "logps/chosen": -977.4869384765625,
913
+ "logps/rejected": -1117.2825927734375,
914
+ "loss": 210782.875,
915
+ "rewards/accuracies": 0.6187499761581421,
916
+ "rewards/chosen": -0.2910914123058319,
917
+ "rewards/margins": 0.06269887834787369,
918
+ "rewards/rejected": -0.3537902534008026,
919
+ "step": 640
920
+ },
921
+ {
922
+ "epoch": 0.52,
923
+ "learning_rate": 2.761321158169134e-06,
924
+ "logits/chosen": 0.4007285237312317,
925
+ "logits/rejected": 0.5178174376487732,
926
+ "logps/chosen": -1064.7994384765625,
927
+ "logps/rejected": -1147.9176025390625,
928
+ "loss": 221734.275,
929
+ "rewards/accuracies": 0.606249988079071,
930
+ "rewards/chosen": -0.32381051778793335,
931
+ "rewards/margins": 0.05276690796017647,
932
+ "rewards/rejected": -0.3765774071216583,
933
+ "step": 650
934
+ },
935
+ {
936
+ "epoch": 0.53,
937
+ "learning_rate": 2.6917975703170466e-06,
938
+ "logits/chosen": 0.4298197329044342,
939
+ "logits/rejected": 0.5217622518539429,
940
+ "logps/chosen": -1024.57568359375,
941
+ "logps/rejected": -1124.2578125,
942
+ "loss": 205664.8875,
943
+ "rewards/accuracies": 0.643750011920929,
944
+ "rewards/chosen": -0.3071673810482025,
945
+ "rewards/margins": 0.05937677621841431,
946
+ "rewards/rejected": -0.3665441572666168,
947
+ "step": 660
948
+ },
949
+ {
950
+ "epoch": 0.54,
951
+ "learning_rate": 2.6221244244890336e-06,
952
+ "logits/chosen": 0.43269839882850647,
953
+ "logits/rejected": 0.5733414888381958,
954
+ "logps/chosen": -1065.0911865234375,
955
+ "logps/rejected": -1168.22705078125,
956
+ "loss": 199175.2125,
957
+ "rewards/accuracies": 0.625,
958
+ "rewards/chosen": -0.36815693974494934,
959
+ "rewards/margins": 0.07322238385677338,
960
+ "rewards/rejected": -0.44137924909591675,
961
+ "step": 670
962
+ },
963
+ {
964
+ "epoch": 0.54,
965
+ "learning_rate": 2.5523560497083927e-06,
966
+ "logits/chosen": 0.48596519231796265,
967
+ "logits/rejected": 0.5404435396194458,
968
+ "logps/chosen": -1015.8805541992188,
969
+ "logps/rejected": -1122.7889404296875,
970
+ "loss": 218825.675,
971
+ "rewards/accuracies": 0.574999988079071,
972
+ "rewards/chosen": -0.3749982416629791,
973
+ "rewards/margins": 0.053486716002225876,
974
+ "rewards/rejected": -0.4284849762916565,
975
+ "step": 680
976
+ },
977
+ {
978
+ "epoch": 0.55,
979
+ "learning_rate": 2.482546849255096e-06,
980
+ "logits/chosen": 0.47428736090660095,
981
+ "logits/rejected": 0.5461061000823975,
982
+ "logps/chosen": -1031.502197265625,
983
+ "logps/rejected": -1122.37451171875,
984
+ "loss": 211378.875,
985
+ "rewards/accuracies": 0.581250011920929,
986
+ "rewards/chosen": -0.3776188790798187,
987
+ "rewards/margins": 0.06876104325056076,
988
+ "rewards/rejected": -0.44637995958328247,
989
+ "step": 690
990
+ },
991
+ {
992
+ "epoch": 0.56,
993
+ "learning_rate": 2.4127512582437486e-06,
994
+ "logits/chosen": 0.5124396085739136,
995
+ "logits/rejected": 0.5212737321853638,
996
+ "logps/chosen": -1060.1768798828125,
997
+ "logps/rejected": -1151.3260498046875,
998
+ "loss": 215612.525,
999
+ "rewards/accuracies": 0.5874999761581421,
1000
+ "rewards/chosen": -0.37767428159713745,
1001
+ "rewards/margins": 0.06161348149180412,
1002
+ "rewards/rejected": -0.43928781151771545,
1003
+ "step": 700
1004
+ },
1005
+ {
1006
+ "epoch": 0.57,
1007
+ "learning_rate": 2.3430237011767166e-06,
1008
+ "logits/chosen": 0.5162058472633362,
1009
+ "logits/rejected": 0.5207083821296692,
1010
+ "logps/chosen": -1115.7698974609375,
1011
+ "logps/rejected": -1238.197998046875,
1012
+ "loss": 207771.0625,
1013
+ "rewards/accuracies": 0.668749988079071,
1014
+ "rewards/chosen": -0.38850077986717224,
1015
+ "rewards/margins": 0.07254378497600555,
1016
+ "rewards/rejected": -0.4610445499420166,
1017
+ "step": 710
1018
+ },
1019
+ {
1020
+ "epoch": 0.58,
1021
+ "learning_rate": 2.2734185495055503e-06,
1022
+ "logits/chosen": 0.5108736753463745,
1023
+ "logits/rejected": 0.565515398979187,
1024
+ "logps/chosen": -1098.859375,
1025
+ "logps/rejected": -1188.5943603515625,
1026
+ "loss": 223396.025,
1027
+ "rewards/accuracies": 0.6187499761581421,
1028
+ "rewards/chosen": -0.4003791809082031,
1029
+ "rewards/margins": 0.05554163455963135,
1030
+ "rewards/rejected": -0.45592084527015686,
1031
+ "step": 720
1032
+ },
1033
+ {
1034
+ "epoch": 0.58,
1035
+ "learning_rate": 2.2039900792337477e-06,
1036
+ "logits/chosen": 0.4386267066001892,
1037
+ "logits/rejected": 0.5263941287994385,
1038
+ "logps/chosen": -1105.8770751953125,
1039
+ "logps/rejected": -1226.1396484375,
1040
+ "loss": 204797.1875,
1041
+ "rewards/accuracies": 0.6187499761581421,
1042
+ "rewards/chosen": -0.39067578315734863,
1043
+ "rewards/margins": 0.07292600721120834,
1044
+ "rewards/rejected": -0.4636017382144928,
1045
+ "step": 730
1046
+ },
1047
+ {
1048
+ "epoch": 0.59,
1049
+ "learning_rate": 2.134792428593971e-06,
1050
+ "logits/chosen": 0.47753873467445374,
1051
+ "logits/rejected": 0.6359479427337646,
1052
+ "logps/chosen": -1050.251220703125,
1053
+ "logps/rejected": -1176.587646484375,
1054
+ "loss": 212099.875,
1055
+ "rewards/accuracies": 0.6000000238418579,
1056
+ "rewards/chosen": -0.3638461232185364,
1057
+ "rewards/margins": 0.0639880821108818,
1058
+ "rewards/rejected": -0.4278342127799988,
1059
+ "step": 740
1060
+ },
1061
+ {
1062
+ "epoch": 0.6,
1063
+ "learning_rate": 2.0658795558326745e-06,
1064
+ "logits/chosen": 0.4289626181125641,
1065
+ "logits/rejected": 0.6244598031044006,
1066
+ "logps/chosen": -1084.2652587890625,
1067
+ "logps/rejected": -1167.8204345703125,
1068
+ "loss": 231680.575,
1069
+ "rewards/accuracies": 0.581250011920929,
1070
+ "rewards/chosen": -0.3578917384147644,
1071
+ "rewards/margins": 0.04770577698945999,
1072
+ "rewards/rejected": -0.4055975079536438,
1073
+ "step": 750
1074
+ },
1075
+ {
1076
+ "epoch": 0.61,
1077
+ "learning_rate": 1.997305197135089e-06,
1078
+ "logits/chosen": 0.49355369806289673,
1079
+ "logits/rejected": 0.5730773210525513,
1080
+ "logps/chosen": -1045.242919921875,
1081
+ "logps/rejected": -1131.892578125,
1082
+ "loss": 224687.025,
1083
+ "rewards/accuracies": 0.5687500238418579,
1084
+ "rewards/chosen": -0.3424818217754364,
1085
+ "rewards/margins": 0.045786917209625244,
1086
+ "rewards/rejected": -0.38826870918273926,
1087
+ "step": 760
1088
+ },
1089
+ {
1090
+ "epoch": 0.62,
1091
+ "learning_rate": 1.9291228247233607e-06,
1092
+ "logits/chosen": 0.47501808404922485,
1093
+ "logits/rejected": 0.4968188405036926,
1094
+ "logps/chosen": -1079.322998046875,
1095
+ "logps/rejected": -1144.870849609375,
1096
+ "loss": 212316.6,
1097
+ "rewards/accuracies": 0.5874999761581421,
1098
+ "rewards/chosen": -0.3503037691116333,
1099
+ "rewards/margins": 0.06730534881353378,
1100
+ "rewards/rejected": -0.4176091253757477,
1101
+ "step": 770
1102
+ },
1103
+ {
1104
+ "epoch": 0.62,
1105
+ "learning_rate": 1.8613856051605242e-06,
1106
+ "logits/chosen": 0.5001388788223267,
1107
+ "logits/rejected": 0.5170911550521851,
1108
+ "logps/chosen": -1020.9500732421875,
1109
+ "logps/rejected": -1150.51953125,
1110
+ "loss": 202457.1,
1111
+ "rewards/accuracies": 0.643750011920929,
1112
+ "rewards/chosen": -0.3424738347530365,
1113
+ "rewards/margins": 0.07792353630065918,
1114
+ "rewards/rejected": -0.42039743065834045,
1115
+ "step": 780
1116
+ },
1117
+ {
1118
+ "epoch": 0.63,
1119
+ "learning_rate": 1.7941463578928088e-06,
1120
+ "logits/chosen": 0.4762800633907318,
1121
+ "logits/rejected": 0.5605748295783997,
1122
+ "logps/chosen": -1117.17724609375,
1123
+ "logps/rejected": -1314.450927734375,
1124
+ "loss": 206942.75,
1125
+ "rewards/accuracies": 0.6499999761581421,
1126
+ "rewards/chosen": -0.4020279049873352,
1127
+ "rewards/margins": 0.08456540107727051,
1128
+ "rewards/rejected": -0.48659324645996094,
1129
+ "step": 790
1130
+ },
1131
+ {
1132
+ "epoch": 0.64,
1133
+ "learning_rate": 1.7274575140626318e-06,
1134
+ "logits/chosen": 0.49529966711997986,
1135
+ "logits/rejected": 0.5294026136398315,
1136
+ "logps/chosen": -1114.677490234375,
1137
+ "logps/rejected": -1209.361572265625,
1138
+ "loss": 225804.925,
1139
+ "rewards/accuracies": 0.6000000238418579,
1140
+ "rewards/chosen": -0.43371039628982544,
1141
+ "rewards/margins": 0.056928206235170364,
1142
+ "rewards/rejected": -0.4906386435031891,
1143
+ "step": 800
1144
+ },
1145
+ {
1146
+ "epoch": 0.65,
1147
+ "learning_rate": 1.661371075624363e-06,
1148
+ "logits/chosen": 0.49185433983802795,
1149
+ "logits/rejected": 0.6701399683952332,
1150
+ "logps/chosen": -1087.6866455078125,
1151
+ "logps/rejected": -1229.163330078125,
1152
+ "loss": 201459.425,
1153
+ "rewards/accuracies": 0.6187499761581421,
1154
+ "rewards/chosen": -0.39431247115135193,
1155
+ "rewards/margins": 0.08681806921958923,
1156
+ "rewards/rejected": -0.4811305105686188,
1157
+ "step": 810
1158
+ },
1159
+ {
1160
+ "epoch": 0.66,
1161
+ "learning_rate": 1.5959385747947697e-06,
1162
+ "logits/chosen": 0.5120656490325928,
1163
+ "logits/rejected": 0.4559970498085022,
1164
+ "logps/chosen": -1053.1092529296875,
1165
+ "logps/rejected": -1140.869384765625,
1166
+ "loss": 210818.75,
1167
+ "rewards/accuracies": 0.59375,
1168
+ "rewards/chosen": -0.347852885723114,
1169
+ "rewards/margins": 0.0646822601556778,
1170
+ "rewards/rejected": -0.4125351309776306,
1171
+ "step": 820
1172
+ },
1173
+ {
1174
+ "epoch": 0.66,
1175
+ "learning_rate": 1.5312110338697427e-06,
1176
+ "logits/chosen": 0.41452938318252563,
1177
+ "logits/rejected": 0.46513715386390686,
1178
+ "logps/chosen": -1092.23193359375,
1179
+ "logps/rejected": -1171.8577880859375,
1180
+ "loss": 218771.075,
1181
+ "rewards/accuracies": 0.543749988079071,
1182
+ "rewards/chosen": -0.3604506552219391,
1183
+ "rewards/margins": 0.05609416961669922,
1184
+ "rewards/rejected": -0.4165448248386383,
1185
+ "step": 830
1186
+ },
1187
+ {
1188
+ "epoch": 0.67,
1189
+ "learning_rate": 1.467238925438646e-06,
1190
+ "logits/chosen": 0.4627233147621155,
1191
+ "logits/rejected": 0.4965755343437195,
1192
+ "logps/chosen": -1006.0794067382812,
1193
+ "logps/rejected": -1153.6485595703125,
1194
+ "loss": 203602.3125,
1195
+ "rewards/accuracies": 0.643750011920929,
1196
+ "rewards/chosen": -0.33338484168052673,
1197
+ "rewards/margins": 0.07592058181762695,
1198
+ "rewards/rejected": -0.4093054234981537,
1199
+ "step": 840
1200
+ },
1201
+ {
1202
+ "epoch": 0.68,
1203
+ "learning_rate": 1.4040721330273063e-06,
1204
+ "logits/chosen": 0.39880701899528503,
1205
+ "logits/rejected": 0.535913348197937,
1206
+ "logps/chosen": -1021.5321044921875,
1207
+ "logps/rejected": -1165.512451171875,
1208
+ "loss": 201742.7,
1209
+ "rewards/accuracies": 0.606249988079071,
1210
+ "rewards/chosen": -0.34407058358192444,
1211
+ "rewards/margins": 0.07992850244045258,
1212
+ "rewards/rejected": -0.4239990711212158,
1213
+ "step": 850
1214
+ },
1215
+ {
1216
+ "epoch": 0.69,
1217
+ "learning_rate": 1.3417599122003464e-06,
1218
+ "logits/chosen": 0.5031896829605103,
1219
+ "logits/rejected": 0.515099287033081,
1220
+ "logps/chosen": -983.9285278320312,
1221
+ "logps/rejected": -1103.140380859375,
1222
+ "loss": 206493.075,
1223
+ "rewards/accuracies": 0.59375,
1224
+ "rewards/chosen": -0.34144875407218933,
1225
+ "rewards/margins": 0.0711575597524643,
1226
+ "rewards/rejected": -0.4126063287258148,
1227
+ "step": 860
1228
+ },
1229
+ {
1230
+ "epoch": 0.7,
1231
+ "learning_rate": 1.280350852153168e-06,
1232
+ "logits/chosen": 0.44252094626426697,
1233
+ "logits/rejected": 0.5791457891464233,
1234
+ "logps/chosen": -1081.3836669921875,
1235
+ "logps/rejected": -1213.262939453125,
1236
+ "loss": 189591.3,
1237
+ "rewards/accuracies": 0.65625,
1238
+ "rewards/chosen": -0.3546789288520813,
1239
+ "rewards/margins": 0.0893690437078476,
1240
+ "rewards/rejected": -0.4440479874610901,
1241
+ "step": 870
1242
+ },
1243
+ {
1244
+ "epoch": 0.7,
1245
+ "learning_rate": 1.2198928378235717e-06,
1246
+ "logits/chosen": 0.4338250160217285,
1247
+ "logits/rejected": 0.4802178740501404,
1248
+ "logps/chosen": -1062.9171142578125,
1249
+ "logps/rejected": -1110.4857177734375,
1250
+ "loss": 216247.0,
1251
+ "rewards/accuracies": 0.574999988079071,
1252
+ "rewards/chosen": -0.34953612089157104,
1253
+ "rewards/margins": 0.05702279880642891,
1254
+ "rewards/rejected": -0.40655893087387085,
1255
+ "step": 880
1256
+ },
1257
+ {
1258
+ "epoch": 0.71,
1259
+ "learning_rate": 1.160433012552508e-06,
1260
+ "logits/chosen": 0.400134801864624,
1261
+ "logits/rejected": 0.5290421843528748,
1262
+ "logps/chosen": -1135.777587890625,
1263
+ "logps/rejected": -1244.5386962890625,
1264
+ "loss": 197283.85,
1265
+ "rewards/accuracies": 0.6312500238418579,
1266
+ "rewards/chosen": -0.37224066257476807,
1267
+ "rewards/margins": 0.07468557357788086,
1268
+ "rewards/rejected": -0.4469262659549713,
1269
+ "step": 890
1270
+ },
1271
+ {
1272
+ "epoch": 0.72,
1273
+ "learning_rate": 1.1020177413231334e-06,
1274
+ "logits/chosen": 0.44932442903518677,
1275
+ "logits/rejected": 0.4805346131324768,
1276
+ "logps/chosen": -1061.0220947265625,
1277
+ "logps/rejected": -1204.8248291015625,
1278
+ "loss": 196671.75,
1279
+ "rewards/accuracies": 0.6187499761581421,
1280
+ "rewards/chosen": -0.3744879364967346,
1281
+ "rewards/margins": 0.083377406001091,
1282
+ "rewards/rejected": -0.45786529779434204,
1283
+ "step": 900
1284
+ },
1285
+ {
1286
+ "epoch": 0.73,
1287
+ "learning_rate": 1.0446925746067768e-06,
1288
+ "logits/chosen": 0.4635135531425476,
1289
+ "logits/rejected": 0.5277852416038513,
1290
+ "logps/chosen": -1039.820068359375,
1291
+ "logps/rejected": -1171.854248046875,
1292
+ "loss": 214841.3,
1293
+ "rewards/accuracies": 0.543749988079071,
1294
+ "rewards/chosen": -0.37799468636512756,
1295
+ "rewards/margins": 0.06353329867124557,
1296
+ "rewards/rejected": -0.44152799248695374,
1297
+ "step": 910
1298
+ },
1299
+ {
1300
+ "epoch": 0.74,
1301
+ "learning_rate": 9.88502212844063e-07,
1302
+ "logits/chosen": 0.39495375752449036,
1303
+ "logits/rejected": 0.5752851963043213,
1304
+ "logps/chosen": -1163.9210205078125,
1305
+ "logps/rejected": -1246.7451171875,
1306
+ "loss": 213209.55,
1307
+ "rewards/accuracies": 0.6000000238418579,
1308
+ "rewards/chosen": -0.4070931077003479,
1309
+ "rewards/margins": 0.0625622421503067,
1310
+ "rewards/rejected": -0.46965542435646057,
1311
+ "step": 920
1312
+ },
1313
+ {
1314
+ "epoch": 0.74,
1315
+ "learning_rate": 9.334904715888496e-07,
1316
+ "logits/chosen": 0.4672183096408844,
1317
+ "logits/rejected": 0.4841803014278412,
1318
+ "logps/chosen": -1080.6539306640625,
1319
+ "logps/rejected": -1161.1947021484375,
1320
+ "loss": 213822.775,
1321
+ "rewards/accuracies": 0.59375,
1322
+ "rewards/chosen": -0.36890339851379395,
1323
+ "rewards/margins": 0.0629233792424202,
1324
+ "rewards/rejected": -0.43182677030563354,
1325
+ "step": 930
1326
+ },
1327
+ {
1328
+ "epoch": 0.75,
1329
+ "learning_rate": 8.797002473421729e-07,
1330
+ "logits/chosen": 0.416917622089386,
1331
+ "logits/rejected": 0.5258146524429321,
1332
+ "logps/chosen": -1045.6466064453125,
1333
+ "logps/rejected": -1163.968017578125,
1334
+ "loss": 211928.45,
1335
+ "rewards/accuracies": 0.6000000238418579,
1336
+ "rewards/chosen": -0.35156455636024475,
1337
+ "rewards/margins": 0.08083046227693558,
1338
+ "rewards/rejected": -0.43239498138427734,
1339
+ "step": 940
1340
+ },
1341
+ {
1342
+ "epoch": 0.76,
1343
+ "learning_rate": 8.271734841028553e-07,
1344
+ "logits/chosen": 0.420457124710083,
1345
+ "logits/rejected": 0.4374917447566986,
1346
+ "logps/chosen": -1128.7008056640625,
1347
+ "logps/rejected": -1171.2813720703125,
1348
+ "loss": 216145.675,
1349
+ "rewards/accuracies": 0.606249988079071,
1350
+ "rewards/chosen": -0.3744417726993561,
1351
+ "rewards/margins": 0.06268437206745148,
1352
+ "rewards/rejected": -0.43712615966796875,
1353
+ "step": 950
1354
+ },
1355
+ {
1356
+ "epoch": 0.77,
1357
+ "learning_rate": 7.759511406608255e-07,
1358
+ "logits/chosen": 0.46132707595825195,
1359
+ "logits/rejected": 0.4851054251194,
1360
+ "logps/chosen": -1103.238525390625,
1361
+ "logps/rejected": -1212.347412109375,
1362
+ "loss": 200265.0875,
1363
+ "rewards/accuracies": 0.637499988079071,
1364
+ "rewards/chosen": -0.366255521774292,
1365
+ "rewards/margins": 0.07792924344539642,
1366
+ "rewards/rejected": -0.4441847801208496,
1367
+ "step": 960
1368
+ },
1369
+ {
1370
+ "epoch": 0.78,
1371
+ "learning_rate": 7.260731586586983e-07,
1372
+ "logits/chosen": 0.4464428424835205,
1373
+ "logits/rejected": 0.5874670743942261,
1374
+ "logps/chosen": -1099.0264892578125,
1375
+ "logps/rejected": -1171.469482421875,
1376
+ "loss": 225859.3,
1377
+ "rewards/accuracies": 0.581250011920929,
1378
+ "rewards/chosen": -0.39616093039512634,
1379
+ "rewards/margins": 0.05346759408712387,
1380
+ "rewards/rejected": -0.44962865114212036,
1381
+ "step": 970
1382
+ },
1383
+ {
1384
+ "epoch": 0.78,
1385
+ "learning_rate": 6.775784314464717e-07,
1386
+ "logits/chosen": 0.43110424280166626,
1387
+ "logits/rejected": 0.4788452982902527,
1388
+ "logps/chosen": -1050.6573486328125,
1389
+ "logps/rejected": -1155.2049560546875,
1390
+ "loss": 212780.5,
1391
+ "rewards/accuracies": 0.581250011920929,
1392
+ "rewards/chosen": -0.3865646719932556,
1393
+ "rewards/margins": 0.06469560414552689,
1394
+ "rewards/rejected": -0.4512602686882019,
1395
+ "step": 980
1396
+ },
1397
+ {
1398
+ "epoch": 0.79,
1399
+ "learning_rate": 6.305047737536707e-07,
1400
+ "logits/chosen": 0.5205137729644775,
1401
+ "logits/rejected": 0.5293751358985901,
1402
+ "logps/chosen": -1037.9459228515625,
1403
+ "logps/rejected": -1127.3016357421875,
1404
+ "loss": 222843.325,
1405
+ "rewards/accuracies": 0.5249999761581421,
1406
+ "rewards/chosen": -0.4062514305114746,
1407
+ "rewards/margins": 0.05210946872830391,
1408
+ "rewards/rejected": -0.45836085081100464,
1409
+ "step": 990
1410
+ },
1411
+ {
1412
+ "epoch": 0.8,
1413
+ "learning_rate": 5.848888922025553e-07,
1414
+ "logits/chosen": 0.4663293957710266,
1415
+ "logits/rejected": 0.6661432981491089,
1416
+ "logps/chosen": -1096.0164794921875,
1417
+ "logps/rejected": -1229.1871337890625,
1418
+ "loss": 206238.925,
1419
+ "rewards/accuracies": 0.574999988079071,
1420
+ "rewards/chosen": -0.42111706733703613,
1421
+ "rewards/margins": 0.07931281626224518,
1422
+ "rewards/rejected": -0.5004298686981201,
1423
+ "step": 1000
1424
+ },
1425
+ {
1426
+ "epoch": 0.81,
1427
+ "learning_rate": 5.407663566854008e-07,
1428
+ "logits/chosen": 0.40173426270484924,
1429
+ "logits/rejected": 0.45892244577407837,
1430
+ "logps/chosen": -1116.3072509765625,
1431
+ "logps/rejected": -1206.285400390625,
1432
+ "loss": 219580.35,
1433
+ "rewards/accuracies": 0.574999988079071,
1434
+ "rewards/chosen": -0.4157632887363434,
1435
+ "rewards/margins": 0.052784573286771774,
1436
+ "rewards/rejected": -0.46854791045188904,
1437
+ "step": 1010
1438
+ },
1439
+ {
1440
+ "epoch": 0.82,
1441
+ "learning_rate": 4.981715726281666e-07,
1442
+ "logits/chosen": 0.44244900345802307,
1443
+ "logits/rejected": 0.5696120262145996,
1444
+ "logps/chosen": -1154.7816162109375,
1445
+ "logps/rejected": -1264.686279296875,
1446
+ "loss": 231070.225,
1447
+ "rewards/accuracies": 0.581250011920929,
1448
+ "rewards/chosen": -0.4453579783439636,
1449
+ "rewards/margins": 0.07253928482532501,
1450
+ "rewards/rejected": -0.5178972482681274,
1451
+ "step": 1020
1452
+ },
1453
+ {
1454
+ "epoch": 0.82,
1455
+ "learning_rate": 4.5713775416217884e-07,
1456
+ "logits/chosen": 0.4136514663696289,
1457
+ "logits/rejected": 0.5045232772827148,
1458
+ "logps/chosen": -1170.0999755859375,
1459
+ "logps/rejected": -1308.028076171875,
1460
+ "loss": 206063.5,
1461
+ "rewards/accuracies": 0.637499988079071,
1462
+ "rewards/chosen": -0.4466446340084076,
1463
+ "rewards/margins": 0.0747566744685173,
1464
+ "rewards/rejected": -0.5214013457298279,
1465
+ "step": 1030
1466
+ },
1467
+ {
1468
+ "epoch": 0.83,
1469
+ "learning_rate": 4.1769689822475147e-07,
1470
+ "logits/chosen": 0.4421358108520508,
1471
+ "logits/rejected": 0.5318270921707153,
1472
+ "logps/chosen": -1143.253662109375,
1473
+ "logps/rejected": -1185.346923828125,
1474
+ "loss": 237069.0,
1475
+ "rewards/accuracies": 0.518750011920929,
1476
+ "rewards/chosen": -0.42274078726768494,
1477
+ "rewards/margins": 0.036188818514347076,
1478
+ "rewards/rejected": -0.4589295983314514,
1479
+ "step": 1040
1480
+ },
1481
+ {
1482
+ "epoch": 0.84,
1483
+ "learning_rate": 3.798797596089351e-07,
1484
+ "logits/chosen": 0.44410261511802673,
1485
+ "logits/rejected": 0.4736444354057312,
1486
+ "logps/chosen": -1085.039306640625,
1487
+ "logps/rejected": -1192.0350341796875,
1488
+ "loss": 218777.575,
1489
+ "rewards/accuracies": 0.5874999761581421,
1490
+ "rewards/chosen": -0.38761842250823975,
1491
+ "rewards/margins": 0.05954207852482796,
1492
+ "rewards/rejected": -0.447160542011261,
1493
+ "step": 1050
1494
+ },
1495
+ {
1496
+ "epoch": 0.85,
1497
+ "learning_rate": 3.4371582698185636e-07,
1498
+ "logits/chosen": 0.4166746139526367,
1499
+ "logits/rejected": 0.439802348613739,
1500
+ "logps/chosen": -1119.6839599609375,
1501
+ "logps/rejected": -1236.3206787109375,
1502
+ "loss": 211817.875,
1503
+ "rewards/accuracies": 0.581250011920929,
1504
+ "rewards/chosen": -0.4001990258693695,
1505
+ "rewards/margins": 0.07398800551891327,
1506
+ "rewards/rejected": -0.47418704628944397,
1507
+ "step": 1060
1508
+ },
1509
+ {
1510
+ "epoch": 0.86,
1511
+ "learning_rate": 3.092332998903416e-07,
1512
+ "logits/chosen": 0.3963729441165924,
1513
+ "logits/rejected": 0.597979724407196,
1514
+ "logps/chosen": -1157.351318359375,
1515
+ "logps/rejected": -1275.6259765625,
1516
+ "loss": 217081.7,
1517
+ "rewards/accuracies": 0.59375,
1518
+ "rewards/chosen": -0.430591881275177,
1519
+ "rewards/margins": 0.07166729122400284,
1520
+ "rewards/rejected": -0.5022591352462769,
1521
+ "step": 1070
1522
+ },
1523
+ {
1524
+ "epoch": 0.86,
1525
+ "learning_rate": 2.764590667717562e-07,
1526
+ "logits/chosen": 0.425032377243042,
1527
+ "logits/rejected": 0.5542384386062622,
1528
+ "logps/chosen": -1154.6519775390625,
1529
+ "logps/rejected": -1220.6429443359375,
1530
+ "loss": 218661.075,
1531
+ "rewards/accuracies": 0.574999988079071,
1532
+ "rewards/chosen": -0.3990860879421234,
1533
+ "rewards/margins": 0.058635927736759186,
1534
+ "rewards/rejected": -0.4577220380306244,
1535
+ "step": 1080
1536
+ },
1537
+ {
1538
+ "epoch": 0.87,
1539
+ "learning_rate": 2.454186839872158e-07,
1540
+ "logits/chosen": 0.3945620656013489,
1541
+ "logits/rejected": 0.5660542249679565,
1542
+ "logps/chosen": -1117.621826171875,
1543
+ "logps/rejected": -1229.8087158203125,
1544
+ "loss": 212127.525,
1545
+ "rewards/accuracies": 0.5874999761581421,
1546
+ "rewards/chosen": -0.3903777599334717,
1547
+ "rewards/margins": 0.06458508223295212,
1548
+ "rewards/rejected": -0.4549628794193268,
1549
+ "step": 1090
1550
+ },
1551
+ {
1552
+ "epoch": 0.88,
1553
+ "learning_rate": 2.1613635589349756e-07,
1554
+ "logits/chosen": 0.47034603357315063,
1555
+ "logits/rejected": 0.49569669365882874,
1556
+ "logps/chosen": -1106.299072265625,
1557
+ "logps/rejected": -1187.548583984375,
1558
+ "loss": 206855.3875,
1559
+ "rewards/accuracies": 0.59375,
1560
+ "rewards/chosen": -0.3820030093193054,
1561
+ "rewards/margins": 0.06976772844791412,
1562
+ "rewards/rejected": -0.4517707824707031,
1563
+ "step": 1100
1564
+ },
1565
+ {
1566
+ "epoch": 0.89,
1567
+ "learning_rate": 1.8863491596921745e-07,
1568
+ "logits/chosen": 0.5025256872177124,
1569
+ "logits/rejected": 0.5709724426269531,
1570
+ "logps/chosen": -1141.001953125,
1571
+ "logps/rejected": -1181.572509765625,
1572
+ "loss": 224914.15,
1573
+ "rewards/accuracies": 0.5687500238418579,
1574
+ "rewards/chosen": -0.3916124105453491,
1575
+ "rewards/margins": 0.05519678443670273,
1576
+ "rewards/rejected": -0.44680923223495483,
1577
+ "step": 1110
1578
+ },
1579
+ {
1580
+ "epoch": 0.9,
1581
+ "learning_rate": 1.629358090099639e-07,
1582
+ "logits/chosen": 0.47358012199401855,
1583
+ "logits/rejected": 0.606777548789978,
1584
+ "logps/chosen": -1059.828125,
1585
+ "logps/rejected": -1188.420166015625,
1586
+ "loss": 192208.55,
1587
+ "rewards/accuracies": 0.6499999761581421,
1588
+ "rewards/chosen": -0.37506911158561707,
1589
+ "rewards/margins": 0.08796657621860504,
1590
+ "rewards/rejected": -0.4630356729030609,
1591
+ "step": 1120
1592
+ },
1593
+ {
1594
+ "epoch": 0.9,
1595
+ "learning_rate": 1.3905907440629752e-07,
1596
+ "logits/chosen": 0.42219334840774536,
1597
+ "logits/rejected": 0.5014327764511108,
1598
+ "logps/chosen": -1034.9349365234375,
1599
+ "logps/rejected": -1115.318603515625,
1600
+ "loss": 215431.625,
1601
+ "rewards/accuracies": 0.5874999761581421,
1602
+ "rewards/chosen": -0.3667599558830261,
1603
+ "rewards/margins": 0.05822211503982544,
1604
+ "rewards/rejected": -0.42498207092285156,
1605
+ "step": 1130
1606
+ },
1607
+ {
1608
+ "epoch": 0.91,
1609
+ "learning_rate": 1.1702333051763271e-07,
1610
+ "logits/chosen": 0.4134256839752197,
1611
+ "logits/rejected": 0.5925799608230591,
1612
+ "logps/chosen": -1156.629638671875,
1613
+ "logps/rejected": -1238.770263671875,
1614
+ "loss": 195104.775,
1615
+ "rewards/accuracies": 0.706250011920929,
1616
+ "rewards/chosen": -0.3899100124835968,
1617
+ "rewards/margins": 0.08536665141582489,
1618
+ "rewards/rejected": -0.4752766489982605,
1619
+ "step": 1140
1620
+ },
1621
+ {
1622
+ "epoch": 0.92,
1623
+ "learning_rate": 9.684576015420277e-08,
1624
+ "logits/chosen": 0.46830758452415466,
1625
+ "logits/rejected": 0.4991118311882019,
1626
+ "logps/chosen": -1111.6732177734375,
1627
+ "logps/rejected": -1219.7283935546875,
1628
+ "loss": 222291.7,
1629
+ "rewards/accuracies": 0.6187499761581421,
1630
+ "rewards/chosen": -0.3950137197971344,
1631
+ "rewards/margins": 0.05590743571519852,
1632
+ "rewards/rejected": -0.4509211480617523,
1633
+ "step": 1150
1634
+ },
1635
+ {
1636
+ "epoch": 0.93,
1637
+ "learning_rate": 7.854209717842231e-08,
1638
+ "logits/chosen": 0.36922842264175415,
1639
+ "logits/rejected": 0.5535188913345337,
1640
+ "logps/chosen": -1031.7603759765625,
1641
+ "logps/rejected": -1112.55517578125,
1642
+ "loss": 205871.2625,
1643
+ "rewards/accuracies": 0.59375,
1644
+ "rewards/chosen": -0.3513695299625397,
1645
+ "rewards/margins": 0.06901805102825165,
1646
+ "rewards/rejected": -0.4203875660896301,
1647
+ "step": 1160
1648
+ },
1649
+ {
1650
+ "epoch": 0.94,
1651
+ "learning_rate": 6.212661423609184e-08,
1652
+ "logits/chosen": 0.4105679988861084,
1653
+ "logits/rejected": 0.6474249958992004,
1654
+ "logps/chosen": -1096.859619140625,
1655
+ "logps/rejected": -1189.704833984375,
1656
+ "loss": 206016.2375,
1657
+ "rewards/accuracies": 0.637499988079071,
1658
+ "rewards/chosen": -0.39043164253234863,
1659
+ "rewards/margins": 0.06852512806653976,
1660
+ "rewards/rejected": -0.4589567184448242,
1661
+ "step": 1170
1662
+ },
1663
+ {
1664
+ "epoch": 0.94,
1665
+ "learning_rate": 4.761211162702117e-08,
1666
+ "logits/chosen": 0.45778220891952515,
1667
+ "logits/rejected": 0.5203276872634888,
1668
+ "logps/chosen": -1121.98193359375,
1669
+ "logps/rejected": -1200.22021484375,
1670
+ "loss": 224875.1,
1671
+ "rewards/accuracies": 0.5687500238418579,
1672
+ "rewards/chosen": -0.39086419343948364,
1673
+ "rewards/margins": 0.0531916543841362,
1674
+ "rewards/rejected": -0.44405585527420044,
1675
+ "step": 1180
1676
+ },
1677
+ {
1678
+ "epoch": 0.95,
1679
+ "learning_rate": 3.5009907323737826e-08,
1680
+ "logits/chosen": 0.5016471147537231,
1681
+ "logits/rejected": 0.4805643558502197,
1682
+ "logps/chosen": -1053.159912109375,
1683
+ "logps/rejected": -1147.296630859375,
1684
+ "loss": 210144.475,
1685
+ "rewards/accuracies": 0.581250011920929,
1686
+ "rewards/chosen": -0.36836639046669006,
1687
+ "rewards/margins": 0.064513199031353,
1688
+ "rewards/rejected": -0.43287959694862366,
1689
+ "step": 1190
1690
+ },
1691
+ {
1692
+ "epoch": 0.96,
1693
+ "learning_rate": 2.4329828146074096e-08,
1694
+ "logits/chosen": 0.4541347026824951,
1695
+ "logits/rejected": 0.5225687026977539,
1696
+ "logps/chosen": -1151.3719482421875,
1697
+ "logps/rejected": -1229.773681640625,
1698
+ "loss": 210966.45,
1699
+ "rewards/accuracies": 0.581250011920929,
1700
+ "rewards/chosen": -0.3953285813331604,
1701
+ "rewards/margins": 0.06661631911993027,
1702
+ "rewards/rejected": -0.46194496750831604,
1703
+ "step": 1200
1704
+ },
1705
+ {
1706
+ "epoch": 0.97,
1707
+ "learning_rate": 1.5580202098509078e-08,
1708
+ "logits/chosen": 0.39697402715682983,
1709
+ "logits/rejected": 0.5624326467514038,
1710
+ "logps/chosen": -1162.04833984375,
1711
+ "logps/rejected": -1252.6485595703125,
1712
+ "loss": 209253.3125,
1713
+ "rewards/accuracies": 0.643750011920929,
1714
+ "rewards/chosen": -0.4083925187587738,
1715
+ "rewards/margins": 0.06362131983041763,
1716
+ "rewards/rejected": -0.4720138609409332,
1717
+ "step": 1210
1718
+ },
1719
+ {
1720
+ "epoch": 0.98,
1721
+ "learning_rate": 8.767851876239075e-09,
1722
+ "logits/chosen": 0.523072361946106,
1723
+ "logits/rejected": 0.5265952944755554,
1724
+ "logps/chosen": -1192.2510986328125,
1725
+ "logps/rejected": -1309.878173828125,
1726
+ "loss": 214136.225,
1727
+ "rewards/accuracies": 0.643750011920929,
1728
+ "rewards/chosen": -0.4248216152191162,
1729
+ "rewards/margins": 0.06676444411277771,
1730
+ "rewards/rejected": -0.49158602952957153,
1731
+ "step": 1220
1732
+ },
1733
+ {
1734
+ "epoch": 0.98,
1735
+ "learning_rate": 3.8980895450474455e-09,
1736
+ "logits/chosen": 0.45322924852371216,
1737
+ "logits/rejected": 0.5091090798377991,
1738
+ "logps/chosen": -1112.177490234375,
1739
+ "logps/rejected": -1221.5980224609375,
1740
+ "loss": 211926.15,
1741
+ "rewards/accuracies": 0.581250011920929,
1742
+ "rewards/chosen": -0.39376866817474365,
1743
+ "rewards/margins": 0.06334400177001953,
1744
+ "rewards/rejected": -0.45711269974708557,
1745
+ "step": 1230
1746
+ },
1747
+ {
1748
+ "epoch": 0.99,
1749
+ "learning_rate": 9.747123991141193e-10,
1750
+ "logits/chosen": 0.39570486545562744,
1751
+ "logits/rejected": 0.5836219191551208,
1752
+ "logps/chosen": -1111.6724853515625,
1753
+ "logps/rejected": -1196.974853515625,
1754
+ "loss": 205382.65,
1755
+ "rewards/accuracies": 0.6875,
1756
+ "rewards/chosen": -0.38581687211990356,
1757
+ "rewards/margins": 0.06402162462472916,
1758
+ "rewards/rejected": -0.44983845949172974,
1759
+ "step": 1240
1760
+ },
1761
+ {
1762
+ "epoch": 1.0,
1763
+ "learning_rate": 0.0,
1764
+ "logits/chosen": 0.4326116144657135,
1765
+ "logits/rejected": 0.5975028872489929,
1766
+ "logps/chosen": -1071.5074462890625,
1767
+ "logps/rejected": -1187.085693359375,
1768
+ "loss": 200385.45,
1769
+ "rewards/accuracies": 0.625,
1770
+ "rewards/chosen": -0.3765797019004822,
1771
+ "rewards/margins": 0.08358252048492432,
1772
+ "rewards/rejected": -0.4601622223854065,
1773
+ "step": 1250
1774
+ },
1775
+ {
1776
+ "epoch": 1.0,
1777
+ "step": 1250,
1778
+ "total_flos": 0.0,
1779
+ "train_loss": 219325.9635,
1780
+ "train_runtime": 11890.0433,
1781
+ "train_samples_per_second": 1.682,
1782
+ "train_steps_per_second": 0.105
1783
+ }
1784
+ ],
1785
+ "logging_steps": 10,
1786
+ "max_steps": 1250,
1787
+ "num_input_tokens_seen": 0,
1788
+ "num_train_epochs": 1,
1789
+ "save_steps": 100,
1790
+ "total_flos": 0.0,
1791
+ "train_batch_size": 4,
1792
+ "trial_name": null,
1793
+ "trial_params": null
1794
+ }