BraylonDash commited on
Commit
45ef2c7
1 Parent(s): 1ae2f93

Model save

Browse files
README.md ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ tags:
4
+ - trl
5
+ - dpo
6
+ - generated_from_trainer
7
+ base_model: DUAL-GPO/zephyr-7b-dpo-0k-15k-i1-merged
8
+ model-index:
9
+ - name: zephyr-7b-dpo-15k-30k-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
+ # zephyr-7b-dpo-15k-30k-i2
17
+
18
+ This model is a fine-tuned version of [DUAL-GPO/zephyr-7b-dpo-0k-15k-i1-merged](https://huggingface.co/DUAL-GPO/zephyr-7b-dpo-0k-15k-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: 2
39
+ - eval_batch_size: 2
40
+ - seed: 42
41
+ - distributed_type: multi-GPU
42
+ - num_devices: 3
43
+ - gradient_accumulation_steps: 2
44
+ - total_train_batch_size: 12
45
+ - total_eval_batch_size: 6
46
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
+ - lr_scheduler_type: cosine
48
+ - lr_scheduler_warmup_ratio: 0.1
49
+ - num_epochs: 1
50
+
51
+ ### Training results
52
+
53
+
54
+
55
+ ### Framework versions
56
+
57
+ - PEFT 0.7.1
58
+ - Transformers 4.36.2
59
+ - Pytorch 2.1.2
60
+ - Datasets 2.14.6
61
+ - Tokenizers 0.15.2
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fc00fd4ba2f084847aa36e7924a4580b4df8d83c007d8cdcb7f52eaaa0562329
3
  size 671150064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:52afbe1ea0ff445dea85e6cc8fb6ead54d8433699dc6a1bd81b9a14136dd52e8
3
  size 671150064
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "train_loss": 0.5400541301727295,
4
+ "train_runtime": 12749.1866,
5
+ "train_samples": 15000,
6
+ "train_samples_per_second": 1.177,
7
+ "train_steps_per_second": 0.098
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
+ 2024-09-19T18:44:50,codecarbon,4ab284e2-06e6-4043-95f6-4fcc134729f1,12749.192444562912,0.006553095341025947,5.140008176612484e-07,42.5,722.254,188.74309015274048,0.15050688559909658,1.9414467976320111,0.6646692117703117,2.7566228950014167,Canada,CAN,quebec,,,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,-71.2,46.8,503.3149070739746,machine,N,1.0
runs/Sep19_15-11-17_gpu4-119-5/events.out.tfevents.1726722741.gpu4-119-5.231915.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b0fb848b91e8110bc2452c111a832d7773032acc866fd80d94053eb93e801648
3
- size 81845
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7cf2bc7cc053a8443ba2ec00dcc198ea3aecc4ed487a608462dae0ea7ba264d9
3
+ size 82833
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "train_loss": 0.5400541301727295,
4
+ "train_runtime": 12749.1866,
5
+ "train_samples": 15000,
6
+ "train_samples_per_second": 1.177,
7
+ "train_steps_per_second": 0.098
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": -1.9630206823349,
15
+ "logits/rejected": -2.162198066711426,
16
+ "logps/chosen": -401.9725341796875,
17
+ "logps/rejected": -476.2349853515625,
18
+ "loss": 0.6931,
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": -1.7936502695083618,
29
+ "logits/rejected": -1.8527445793151855,
30
+ "logps/chosen": -465.740234375,
31
+ "logps/rejected": -598.78759765625,
32
+ "loss": 0.6927,
33
+ "rewards/accuracies": 0.4722222089767456,
34
+ "rewards/chosen": 0.0014809627318754792,
35
+ "rewards/margins": 0.004310874734073877,
36
+ "rewards/rejected": -0.0028299118857830763,
37
+ "step": 10
38
+ },
39
+ {
40
+ "epoch": 0.02,
41
+ "learning_rate": 8.000000000000001e-07,
42
+ "logits/chosen": -1.9694637060165405,
43
+ "logits/rejected": -1.6243550777435303,
44
+ "logps/chosen": -553.727783203125,
45
+ "logps/rejected": -633.4371337890625,
46
+ "loss": 0.6877,
47
+ "rewards/accuracies": 0.5,
48
+ "rewards/chosen": -0.01299777440726757,
49
+ "rewards/margins": 0.013292083516716957,
50
+ "rewards/rejected": -0.026289859786629677,
51
+ "step": 20
52
+ },
53
+ {
54
+ "epoch": 0.02,
55
+ "learning_rate": 1.2000000000000002e-06,
56
+ "logits/chosen": -1.95724618434906,
57
+ "logits/rejected": -2.026108980178833,
58
+ "logps/chosen": -514.1293334960938,
59
+ "logps/rejected": -665.3492431640625,
60
+ "loss": 0.6636,
61
+ "rewards/accuracies": 0.6499999761581421,
62
+ "rewards/chosen": -0.1179969534277916,
63
+ "rewards/margins": 0.06291333585977554,
64
+ "rewards/rejected": -0.18091025948524475,
65
+ "step": 30
66
+ },
67
+ {
68
+ "epoch": 0.03,
69
+ "learning_rate": 1.6000000000000001e-06,
70
+ "logits/chosen": -1.6502164602279663,
71
+ "logits/rejected": -1.3826799392700195,
72
+ "logps/chosen": -581.6859741210938,
73
+ "logps/rejected": -672.312744140625,
74
+ "loss": 0.6306,
75
+ "rewards/accuracies": 0.6499999761581421,
76
+ "rewards/chosen": -0.30207711458206177,
77
+ "rewards/margins": 0.21752643585205078,
78
+ "rewards/rejected": -0.5196035504341125,
79
+ "step": 40
80
+ },
81
+ {
82
+ "epoch": 0.04,
83
+ "learning_rate": 2.0000000000000003e-06,
84
+ "logits/chosen": -1.8586839437484741,
85
+ "logits/rejected": -1.6029729843139648,
86
+ "logps/chosen": -581.8325805664062,
87
+ "logps/rejected": -661.6460571289062,
88
+ "loss": 0.5985,
89
+ "rewards/accuracies": 0.6499999761581421,
90
+ "rewards/chosen": -0.5655128955841064,
91
+ "rewards/margins": 0.35827767848968506,
92
+ "rewards/rejected": -0.9237905740737915,
93
+ "step": 50
94
+ },
95
+ {
96
+ "epoch": 0.05,
97
+ "learning_rate": 2.4000000000000003e-06,
98
+ "logits/chosen": -1.830959677696228,
99
+ "logits/rejected": -1.677363634109497,
100
+ "logps/chosen": -705.7425537109375,
101
+ "logps/rejected": -838.0233154296875,
102
+ "loss": 0.6206,
103
+ "rewards/accuracies": 0.6000000238418579,
104
+ "rewards/chosen": -1.1965574026107788,
105
+ "rewards/margins": 0.3305513560771942,
106
+ "rewards/rejected": -1.5271086692810059,
107
+ "step": 60
108
+ },
109
+ {
110
+ "epoch": 0.06,
111
+ "learning_rate": 2.8000000000000003e-06,
112
+ "logits/chosen": -1.5858336687088013,
113
+ "logits/rejected": -1.5155165195465088,
114
+ "logps/chosen": -586.5440673828125,
115
+ "logps/rejected": -717.4639282226562,
116
+ "loss": 0.6001,
117
+ "rewards/accuracies": 0.675000011920929,
118
+ "rewards/chosen": -0.7764847874641418,
119
+ "rewards/margins": 0.5009865164756775,
120
+ "rewards/rejected": -1.2774713039398193,
121
+ "step": 70
122
+ },
123
+ {
124
+ "epoch": 0.06,
125
+ "learning_rate": 3.2000000000000003e-06,
126
+ "logits/chosen": -1.7760032415390015,
127
+ "logits/rejected": -1.6308609247207642,
128
+ "logps/chosen": -639.7242431640625,
129
+ "logps/rejected": -749.7010498046875,
130
+ "loss": 0.5326,
131
+ "rewards/accuracies": 0.675000011920929,
132
+ "rewards/chosen": -0.8676024675369263,
133
+ "rewards/margins": 0.5284737348556519,
134
+ "rewards/rejected": -1.3960762023925781,
135
+ "step": 80
136
+ },
137
+ {
138
+ "epoch": 0.07,
139
+ "learning_rate": 3.6000000000000003e-06,
140
+ "logits/chosen": -1.7125171422958374,
141
+ "logits/rejected": -1.546307921409607,
142
+ "logps/chosen": -815.2091064453125,
143
+ "logps/rejected": -867.2105712890625,
144
+ "loss": 0.5442,
145
+ "rewards/accuracies": 0.6499999761581421,
146
+ "rewards/chosen": -2.1319868564605713,
147
+ "rewards/margins": 0.48750001192092896,
148
+ "rewards/rejected": -2.6194865703582764,
149
+ "step": 90
150
+ },
151
+ {
152
+ "epoch": 0.08,
153
+ "learning_rate": 4.000000000000001e-06,
154
+ "logits/chosen": -1.6376460790634155,
155
+ "logits/rejected": -1.7543089389801025,
156
+ "logps/chosen": -840.8780517578125,
157
+ "logps/rejected": -872.4625244140625,
158
+ "loss": 0.7056,
159
+ "rewards/accuracies": 0.550000011920929,
160
+ "rewards/chosen": -3.0078701972961426,
161
+ "rewards/margins": 0.050300419330596924,
162
+ "rewards/rejected": -3.0581705570220947,
163
+ "step": 100
164
+ },
165
+ {
166
+ "epoch": 0.09,
167
+ "learning_rate": 4.4e-06,
168
+ "logits/chosen": -1.6319936513900757,
169
+ "logits/rejected": -1.5396147966384888,
170
+ "logps/chosen": -589.26513671875,
171
+ "logps/rejected": -693.470458984375,
172
+ "loss": 0.5643,
173
+ "rewards/accuracies": 0.550000011920929,
174
+ "rewards/chosen": -1.3587239980697632,
175
+ "rewards/margins": 0.4278411865234375,
176
+ "rewards/rejected": -1.7865655422210693,
177
+ "step": 110
178
+ },
179
+ {
180
+ "epoch": 0.1,
181
+ "learning_rate": 4.800000000000001e-06,
182
+ "logits/chosen": -1.8605772256851196,
183
+ "logits/rejected": -1.6579277515411377,
184
+ "logps/chosen": -635.4324340820312,
185
+ "logps/rejected": -726.6505126953125,
186
+ "loss": 0.5762,
187
+ "rewards/accuracies": 0.625,
188
+ "rewards/chosen": -1.2179909944534302,
189
+ "rewards/margins": 0.410940557718277,
190
+ "rewards/rejected": -1.6289316415786743,
191
+ "step": 120
192
+ },
193
+ {
194
+ "epoch": 0.1,
195
+ "learning_rate": 4.999756310023261e-06,
196
+ "logits/chosen": -1.6890239715576172,
197
+ "logits/rejected": -1.708566427230835,
198
+ "logps/chosen": -628.5355224609375,
199
+ "logps/rejected": -787.552490234375,
200
+ "loss": 0.5259,
201
+ "rewards/accuracies": 0.6499999761581421,
202
+ "rewards/chosen": -1.2577191591262817,
203
+ "rewards/margins": 0.6898480653762817,
204
+ "rewards/rejected": -1.9475672245025635,
205
+ "step": 130
206
+ },
207
+ {
208
+ "epoch": 0.11,
209
+ "learning_rate": 4.997807075247147e-06,
210
+ "logits/chosen": -1.801041603088379,
211
+ "logits/rejected": -1.5225584506988525,
212
+ "logps/chosen": -595.2302856445312,
213
+ "logps/rejected": -674.1448974609375,
214
+ "loss": 0.5767,
215
+ "rewards/accuracies": 0.42500001192092896,
216
+ "rewards/chosen": -0.8753921389579773,
217
+ "rewards/margins": 0.308530330657959,
218
+ "rewards/rejected": -1.183922529220581,
219
+ "step": 140
220
+ },
221
+ {
222
+ "epoch": 0.12,
223
+ "learning_rate": 4.993910125649561e-06,
224
+ "logits/chosen": -1.4428200721740723,
225
+ "logits/rejected": -1.3668603897094727,
226
+ "logps/chosen": -705.331298828125,
227
+ "logps/rejected": -784.1256103515625,
228
+ "loss": 0.5467,
229
+ "rewards/accuracies": 0.574999988079071,
230
+ "rewards/chosen": -1.4335416555404663,
231
+ "rewards/margins": 0.4211806356906891,
232
+ "rewards/rejected": -1.8547226190567017,
233
+ "step": 150
234
+ },
235
+ {
236
+ "epoch": 0.13,
237
+ "learning_rate": 4.988068499954578e-06,
238
+ "logits/chosen": -1.7208309173583984,
239
+ "logits/rejected": -1.5595357418060303,
240
+ "logps/chosen": -798.42236328125,
241
+ "logps/rejected": -894.1146240234375,
242
+ "loss": 0.5702,
243
+ "rewards/accuracies": 0.6000000238418579,
244
+ "rewards/chosen": -2.107240676879883,
245
+ "rewards/margins": 0.4062272012233734,
246
+ "rewards/rejected": -2.513467788696289,
247
+ "step": 160
248
+ },
249
+ {
250
+ "epoch": 0.14,
251
+ "learning_rate": 4.980286753286196e-06,
252
+ "logits/chosen": -1.653246283531189,
253
+ "logits/rejected": -1.5499178171157837,
254
+ "logps/chosen": -733.5491943359375,
255
+ "logps/rejected": -826.2314453125,
256
+ "loss": 0.639,
257
+ "rewards/accuracies": 0.6000000238418579,
258
+ "rewards/chosen": -2.0218682289123535,
259
+ "rewards/margins": 0.3602715730667114,
260
+ "rewards/rejected": -2.3821396827697754,
261
+ "step": 170
262
+ },
263
+ {
264
+ "epoch": 0.14,
265
+ "learning_rate": 4.970570953616383e-06,
266
+ "logits/chosen": -1.844082236289978,
267
+ "logits/rejected": -1.434446930885315,
268
+ "logps/chosen": -690.146728515625,
269
+ "logps/rejected": -739.8771362304688,
270
+ "loss": 0.6021,
271
+ "rewards/accuracies": 0.574999988079071,
272
+ "rewards/chosen": -1.4458850622177124,
273
+ "rewards/margins": 0.38214051723480225,
274
+ "rewards/rejected": -1.828025460243225,
275
+ "step": 180
276
+ },
277
+ {
278
+ "epoch": 0.15,
279
+ "learning_rate": 4.958928677033465e-06,
280
+ "logits/chosen": -1.8028007745742798,
281
+ "logits/rejected": -1.5488333702087402,
282
+ "logps/chosen": -849.9052734375,
283
+ "logps/rejected": -1006.4519653320312,
284
+ "loss": 0.5145,
285
+ "rewards/accuracies": 0.7749999761581421,
286
+ "rewards/chosen": -2.3674304485321045,
287
+ "rewards/margins": 0.8914012908935547,
288
+ "rewards/rejected": -3.258831739425659,
289
+ "step": 190
290
+ },
291
+ {
292
+ "epoch": 0.16,
293
+ "learning_rate": 4.9453690018345144e-06,
294
+ "logits/chosen": -1.8077857494354248,
295
+ "logits/rejected": -1.7000172138214111,
296
+ "logps/chosen": -915.9679565429688,
297
+ "logps/rejected": -996.1795654296875,
298
+ "loss": 0.6229,
299
+ "rewards/accuracies": 0.574999988079071,
300
+ "rewards/chosen": -2.8740787506103516,
301
+ "rewards/margins": 0.5032685995101929,
302
+ "rewards/rejected": -3.377347469329834,
303
+ "step": 200
304
+ },
305
+ {
306
+ "epoch": 0.17,
307
+ "learning_rate": 4.9299025014463665e-06,
308
+ "logits/chosen": -1.6895710229873657,
309
+ "logits/rejected": -1.4762156009674072,
310
+ "logps/chosen": -723.7333374023438,
311
+ "logps/rejected": -790.6909790039062,
312
+ "loss": 0.5732,
313
+ "rewards/accuracies": 0.625,
314
+ "rewards/chosen": -1.8080753087997437,
315
+ "rewards/margins": 0.33258000016212463,
316
+ "rewards/rejected": -2.140655040740967,
317
+ "step": 210
318
+ },
319
+ {
320
+ "epoch": 0.18,
321
+ "learning_rate": 4.912541236180779e-06,
322
+ "logits/chosen": -1.7490249872207642,
323
+ "logits/rejected": -1.6070995330810547,
324
+ "logps/chosen": -712.7299194335938,
325
+ "logps/rejected": -795.3870849609375,
326
+ "loss": 0.574,
327
+ "rewards/accuracies": 0.625,
328
+ "rewards/chosen": -1.5378398895263672,
329
+ "rewards/margins": 0.3792051076889038,
330
+ "rewards/rejected": -1.917044997215271,
331
+ "step": 220
332
+ },
333
+ {
334
+ "epoch": 0.18,
335
+ "learning_rate": 4.893298743830168e-06,
336
+ "logits/chosen": -1.900240182876587,
337
+ "logits/rejected": -1.6148269176483154,
338
+ "logps/chosen": -722.8997192382812,
339
+ "logps/rejected": -808.047607421875,
340
+ "loss": 0.5498,
341
+ "rewards/accuracies": 0.625,
342
+ "rewards/chosen": -1.4032080173492432,
343
+ "rewards/margins": 0.5880616903305054,
344
+ "rewards/rejected": -1.9912694692611694,
345
+ "step": 230
346
+ },
347
+ {
348
+ "epoch": 0.19,
349
+ "learning_rate": 4.8721900291112415e-06,
350
+ "logits/chosen": -1.7972081899642944,
351
+ "logits/rejected": -1.8194860219955444,
352
+ "logps/chosen": -734.0165405273438,
353
+ "logps/rejected": -861.5304565429688,
354
+ "loss": 0.5786,
355
+ "rewards/accuracies": 0.675000011920929,
356
+ "rewards/chosen": -1.638071060180664,
357
+ "rewards/margins": 0.5583940148353577,
358
+ "rewards/rejected": -2.196465015411377,
359
+ "step": 240
360
+ },
361
+ {
362
+ "epoch": 0.2,
363
+ "learning_rate": 4.849231551964771e-06,
364
+ "logits/chosen": -1.8508936166763306,
365
+ "logits/rejected": -1.80252206325531,
366
+ "logps/chosen": -721.9108276367188,
367
+ "logps/rejected": -872.9285888671875,
368
+ "loss": 0.5494,
369
+ "rewards/accuracies": 0.625,
370
+ "rewards/chosen": -1.3129396438598633,
371
+ "rewards/margins": 0.5180081129074097,
372
+ "rewards/rejected": -1.8309476375579834,
373
+ "step": 250
374
+ },
375
+ {
376
+ "epoch": 0.21,
377
+ "learning_rate": 4.824441214720629e-06,
378
+ "logits/chosen": -1.7704343795776367,
379
+ "logits/rejected": -1.3905961513519287,
380
+ "logps/chosen": -711.2572631835938,
381
+ "logps/rejected": -867.6805419921875,
382
+ "loss": 0.5413,
383
+ "rewards/accuracies": 0.7250000238418579,
384
+ "rewards/chosen": -1.5769939422607422,
385
+ "rewards/margins": 0.8006278276443481,
386
+ "rewards/rejected": -2.37762188911438,
387
+ "step": 260
388
+ },
389
+ {
390
+ "epoch": 0.22,
391
+ "learning_rate": 4.7978383481380865e-06,
392
+ "logits/chosen": -1.6354033946990967,
393
+ "logits/rejected": -1.5733468532562256,
394
+ "logps/chosen": -834.4849853515625,
395
+ "logps/rejected": -1068.08154296875,
396
+ "loss": 0.5731,
397
+ "rewards/accuracies": 0.7250000238418579,
398
+ "rewards/chosen": -2.7554314136505127,
399
+ "rewards/margins": 1.313236117362976,
400
+ "rewards/rejected": -4.068667411804199,
401
+ "step": 270
402
+ },
403
+ {
404
+ "epoch": 0.22,
405
+ "learning_rate": 4.769443696332272e-06,
406
+ "logits/chosen": -1.9346412420272827,
407
+ "logits/rejected": -1.8029006719589233,
408
+ "logps/chosen": -814.7976684570312,
409
+ "logps/rejected": -931.1267700195312,
410
+ "loss": 0.5191,
411
+ "rewards/accuracies": 0.5249999761581421,
412
+ "rewards/chosen": -2.9693081378936768,
413
+ "rewards/margins": 0.6553217172622681,
414
+ "rewards/rejected": -3.624629497528076,
415
+ "step": 280
416
+ },
417
+ {
418
+ "epoch": 0.23,
419
+ "learning_rate": 4.7392794005985324e-06,
420
+ "logits/chosen": -1.9160423278808594,
421
+ "logits/rejected": -1.6570581197738647,
422
+ "logps/chosen": -743.0398559570312,
423
+ "logps/rejected": -913.5885009765625,
424
+ "loss": 0.5361,
425
+ "rewards/accuracies": 0.75,
426
+ "rewards/chosen": -2.2374563217163086,
427
+ "rewards/margins": 1.1544185876846313,
428
+ "rewards/rejected": -3.3918750286102295,
429
+ "step": 290
430
+ },
431
+ {
432
+ "epoch": 0.24,
433
+ "learning_rate": 4.707368982147318e-06,
434
+ "logits/chosen": -1.6389538049697876,
435
+ "logits/rejected": -1.597599744796753,
436
+ "logps/chosen": -877.5833740234375,
437
+ "logps/rejected": -1009.7105712890625,
438
+ "loss": 0.6373,
439
+ "rewards/accuracies": 0.574999988079071,
440
+ "rewards/chosen": -2.7767863273620605,
441
+ "rewards/margins": 0.527004599571228,
442
+ "rewards/rejected": -3.303791046142578,
443
+ "step": 300
444
+ },
445
+ {
446
+ "epoch": 0.25,
447
+ "learning_rate": 4.673737323763048e-06,
448
+ "logits/chosen": -1.6859241724014282,
449
+ "logits/rejected": -1.5517866611480713,
450
+ "logps/chosen": -697.40771484375,
451
+ "logps/rejected": -884.6589965820312,
452
+ "loss": 0.5183,
453
+ "rewards/accuracies": 0.699999988079071,
454
+ "rewards/chosen": -1.7637040615081787,
455
+ "rewards/margins": 0.8891094923019409,
456
+ "rewards/rejected": -2.652813673019409,
457
+ "step": 310
458
+ },
459
+ {
460
+ "epoch": 0.26,
461
+ "learning_rate": 4.638410650401267e-06,
462
+ "logits/chosen": -1.7035640478134155,
463
+ "logits/rejected": -1.3526010513305664,
464
+ "logps/chosen": -746.2251586914062,
465
+ "logps/rejected": -828.48779296875,
466
+ "loss": 0.5582,
467
+ "rewards/accuracies": 0.675000011920929,
468
+ "rewards/chosen": -1.4476099014282227,
469
+ "rewards/margins": 0.6334011554718018,
470
+ "rewards/rejected": -2.0810108184814453,
471
+ "step": 320
472
+ },
473
+ {
474
+ "epoch": 0.26,
475
+ "learning_rate": 4.601416508739211e-06,
476
+ "logits/chosen": -1.6877880096435547,
477
+ "logits/rejected": -1.5339024066925049,
478
+ "logps/chosen": -686.0667724609375,
479
+ "logps/rejected": -927.6800537109375,
480
+ "loss": 0.5501,
481
+ "rewards/accuracies": 0.625,
482
+ "rewards/chosen": -1.6149232387542725,
483
+ "rewards/margins": 0.9856560826301575,
484
+ "rewards/rejected": -2.600579261779785,
485
+ "step": 330
486
+ },
487
+ {
488
+ "epoch": 0.27,
489
+ "learning_rate": 4.562783745695738e-06,
490
+ "logits/chosen": -1.4515044689178467,
491
+ "logits/rejected": -1.5230369567871094,
492
+ "logps/chosen": -600.0208740234375,
493
+ "logps/rejected": -834.6480712890625,
494
+ "loss": 0.563,
495
+ "rewards/accuracies": 0.7250000238418579,
496
+ "rewards/chosen": -1.0641671419143677,
497
+ "rewards/margins": 0.8571918606758118,
498
+ "rewards/rejected": -1.9213588237762451,
499
+ "step": 340
500
+ },
501
+ {
502
+ "epoch": 0.28,
503
+ "learning_rate": 4.522542485937369e-06,
504
+ "logits/chosen": -1.5642414093017578,
505
+ "logits/rejected": -1.4102091789245605,
506
+ "logps/chosen": -623.0133666992188,
507
+ "logps/rejected": -696.9454345703125,
508
+ "loss": 0.5581,
509
+ "rewards/accuracies": 0.6000000238418579,
510
+ "rewards/chosen": -0.6556235551834106,
511
+ "rewards/margins": 0.5390894412994385,
512
+ "rewards/rejected": -1.1947131156921387,
513
+ "step": 350
514
+ },
515
+ {
516
+ "epoch": 0.29,
517
+ "learning_rate": 4.4807241083879774e-06,
518
+ "logits/chosen": -1.4413741827011108,
519
+ "logits/rejected": -1.3161842823028564,
520
+ "logps/chosen": -637.3552856445312,
521
+ "logps/rejected": -746.97509765625,
522
+ "loss": 0.6226,
523
+ "rewards/accuracies": 0.5249999761581421,
524
+ "rewards/chosen": -0.7561216950416565,
525
+ "rewards/margins": 0.5896363854408264,
526
+ "rewards/rejected": -1.345758080482483,
527
+ "step": 360
528
+ },
529
+ {
530
+ "epoch": 0.3,
531
+ "learning_rate": 4.437361221760449e-06,
532
+ "logits/chosen": -1.4904801845550537,
533
+ "logits/rejected": -1.364058017730713,
534
+ "logps/chosen": -606.2081298828125,
535
+ "logps/rejected": -696.7344970703125,
536
+ "loss": 0.5533,
537
+ "rewards/accuracies": 0.4749999940395355,
538
+ "rewards/chosen": -0.5675913691520691,
539
+ "rewards/margins": 0.32215815782546997,
540
+ "rewards/rejected": -0.8897494077682495,
541
+ "step": 370
542
+ },
543
+ {
544
+ "epoch": 0.3,
545
+ "learning_rate": 4.3924876391293915e-06,
546
+ "logits/chosen": -1.5729104280471802,
547
+ "logits/rejected": -1.5222874879837036,
548
+ "logps/chosen": -599.6661987304688,
549
+ "logps/rejected": -789.137451171875,
550
+ "loss": 0.5297,
551
+ "rewards/accuracies": 0.6499999761581421,
552
+ "rewards/chosen": -0.6211053729057312,
553
+ "rewards/margins": 0.7193719148635864,
554
+ "rewards/rejected": -1.3404772281646729,
555
+ "step": 380
556
+ },
557
+ {
558
+ "epoch": 0.31,
559
+ "learning_rate": 4.346138351564711e-06,
560
+ "logits/chosen": -1.6941932439804077,
561
+ "logits/rejected": -1.6446470022201538,
562
+ "logps/chosen": -659.5060424804688,
563
+ "logps/rejected": -771.0955810546875,
564
+ "loss": 0.5252,
565
+ "rewards/accuracies": 0.625,
566
+ "rewards/chosen": -0.9060527086257935,
567
+ "rewards/margins": 0.619795024394989,
568
+ "rewards/rejected": -1.5258477926254272,
569
+ "step": 390
570
+ },
571
+ {
572
+ "epoch": 0.32,
573
+ "learning_rate": 4.2983495008466285e-06,
574
+ "logits/chosen": -1.8499412536621094,
575
+ "logits/rejected": -1.6332486867904663,
576
+ "logps/chosen": -627.735595703125,
577
+ "logps/rejected": -768.0028076171875,
578
+ "loss": 0.537,
579
+ "rewards/accuracies": 0.6499999761581421,
580
+ "rewards/chosen": -0.6158324480056763,
581
+ "rewards/margins": 0.8914637565612793,
582
+ "rewards/rejected": -1.5072962045669556,
583
+ "step": 400
584
+ },
585
+ {
586
+ "epoch": 0.33,
587
+ "learning_rate": 4.249158351283414e-06,
588
+ "logits/chosen": -1.7273086309432983,
589
+ "logits/rejected": -1.5036022663116455,
590
+ "logps/chosen": -523.8494873046875,
591
+ "logps/rejected": -641.2108154296875,
592
+ "loss": 0.5355,
593
+ "rewards/accuracies": 0.5249999761581421,
594
+ "rewards/chosen": -0.9005772471427917,
595
+ "rewards/margins": 0.5681300163269043,
596
+ "rewards/rejected": -1.4687073230743408,
597
+ "step": 410
598
+ },
599
+ {
600
+ "epoch": 0.34,
601
+ "learning_rate": 4.198603260653792e-06,
602
+ "logits/chosen": -1.7295076847076416,
603
+ "logits/rejected": -1.4331910610198975,
604
+ "logps/chosen": -721.2381591796875,
605
+ "logps/rejected": -866.8610229492188,
606
+ "loss": 0.5564,
607
+ "rewards/accuracies": 0.6499999761581421,
608
+ "rewards/chosen": -1.6372039318084717,
609
+ "rewards/margins": 0.8299415707588196,
610
+ "rewards/rejected": -2.4671454429626465,
611
+ "step": 420
612
+ },
613
+ {
614
+ "epoch": 0.34,
615
+ "learning_rate": 4.146723650296701e-06,
616
+ "logits/chosen": -1.5709054470062256,
617
+ "logits/rejected": -1.5238134860992432,
618
+ "logps/chosen": -833.8419799804688,
619
+ "logps/rejected": -953.95263671875,
620
+ "loss": 0.5362,
621
+ "rewards/accuracies": 0.6499999761581421,
622
+ "rewards/chosen": -2.0778729915618896,
623
+ "rewards/margins": 0.6275758743286133,
624
+ "rewards/rejected": -2.705448865890503,
625
+ "step": 430
626
+ },
627
+ {
628
+ "epoch": 0.35,
629
+ "learning_rate": 4.093559974371725e-06,
630
+ "logits/chosen": -1.8162000179290771,
631
+ "logits/rejected": -1.7265230417251587,
632
+ "logps/chosen": -681.6597900390625,
633
+ "logps/rejected": -798.3497314453125,
634
+ "loss": 0.5828,
635
+ "rewards/accuracies": 0.625,
636
+ "rewards/chosen": -1.323164939880371,
637
+ "rewards/margins": 0.6071309447288513,
638
+ "rewards/rejected": -1.9302961826324463,
639
+ "step": 440
640
+ },
641
+ {
642
+ "epoch": 0.36,
643
+ "learning_rate": 4.039153688314146e-06,
644
+ "logits/chosen": -1.678344964981079,
645
+ "logits/rejected": -1.357090711593628,
646
+ "logps/chosen": -859.0988159179688,
647
+ "logps/rejected": -954.8541259765625,
648
+ "loss": 0.6051,
649
+ "rewards/accuracies": 0.4749999940395355,
650
+ "rewards/chosen": -2.2214248180389404,
651
+ "rewards/margins": 0.560940146446228,
652
+ "rewards/rejected": -2.782365083694458,
653
+ "step": 450
654
+ },
655
+ {
656
+ "epoch": 0.37,
657
+ "learning_rate": 3.983547216509254e-06,
658
+ "logits/chosen": -1.9185292720794678,
659
+ "logits/rejected": -1.641157865524292,
660
+ "logps/chosen": -733.2647094726562,
661
+ "logps/rejected": -789.9527587890625,
662
+ "loss": 0.5611,
663
+ "rewards/accuracies": 0.550000011920929,
664
+ "rewards/chosen": -1.8194106817245483,
665
+ "rewards/margins": 0.43999481201171875,
666
+ "rewards/rejected": -2.2594056129455566,
667
+ "step": 460
668
+ },
669
+ {
670
+ "epoch": 0.38,
671
+ "learning_rate": 3.92678391921108e-06,
672
+ "logits/chosen": -1.435787558555603,
673
+ "logits/rejected": -1.475290060043335,
674
+ "logps/chosen": -730.6599731445312,
675
+ "logps/rejected": -898.69287109375,
676
+ "loss": 0.5107,
677
+ "rewards/accuracies": 0.675000011920929,
678
+ "rewards/chosen": -2.1700119972229004,
679
+ "rewards/margins": 0.8106654286384583,
680
+ "rewards/rejected": -2.980677843093872,
681
+ "step": 470
682
+ },
683
+ {
684
+ "epoch": 0.38,
685
+ "learning_rate": 3.868908058731376e-06,
686
+ "logits/chosen": -1.608912706375122,
687
+ "logits/rejected": -1.413901686668396,
688
+ "logps/chosen": -612.453369140625,
689
+ "logps/rejected": -707.7071533203125,
690
+ "loss": 0.6148,
691
+ "rewards/accuracies": 0.5,
692
+ "rewards/chosen": -1.772383451461792,
693
+ "rewards/margins": 0.5594812631607056,
694
+ "rewards/rejected": -2.331864833831787,
695
+ "step": 480
696
+ },
697
+ {
698
+ "epoch": 0.39,
699
+ "learning_rate": 3.8099647649251984e-06,
700
+ "logits/chosen": -1.6021438837051392,
701
+ "logits/rejected": -1.5742886066436768,
702
+ "logps/chosen": -611.6611938476562,
703
+ "logps/rejected": -834.7572021484375,
704
+ "loss": 0.487,
705
+ "rewards/accuracies": 0.625,
706
+ "rewards/chosen": -1.202609658241272,
707
+ "rewards/margins": 1.1881163120269775,
708
+ "rewards/rejected": -2.390725612640381,
709
+ "step": 490
710
+ },
711
+ {
712
+ "epoch": 0.4,
713
+ "learning_rate": 3.7500000000000005e-06,
714
+ "logits/chosen": -1.6629788875579834,
715
+ "logits/rejected": -1.4489120244979858,
716
+ "logps/chosen": -608.3432006835938,
717
+ "logps/rejected": -702.0739135742188,
718
+ "loss": 0.5747,
719
+ "rewards/accuracies": 0.5249999761581421,
720
+ "rewards/chosen": -0.9137832522392273,
721
+ "rewards/margins": 0.521483302116394,
722
+ "rewards/rejected": -1.435266375541687,
723
+ "step": 500
724
+ },
725
+ {
726
+ "epoch": 0.41,
727
+ "learning_rate": 3.689060522675689e-06,
728
+ "logits/chosen": -1.7527472972869873,
729
+ "logits/rejected": -1.5482290983200073,
730
+ "logps/chosen": -672.4737548828125,
731
+ "logps/rejected": -798.4075317382812,
732
+ "loss": 0.497,
733
+ "rewards/accuracies": 0.7250000238418579,
734
+ "rewards/chosen": -1.3834949731826782,
735
+ "rewards/margins": 0.8907042741775513,
736
+ "rewards/rejected": -2.2741992473602295,
737
+ "step": 510
738
+ },
739
+ {
740
+ "epoch": 0.42,
741
+ "learning_rate": 3.627193851723577e-06,
742
+ "logits/chosen": -1.791130781173706,
743
+ "logits/rejected": -1.467633843421936,
744
+ "logps/chosen": -787.7621459960938,
745
+ "logps/rejected": -947.30419921875,
746
+ "loss": 0.5292,
747
+ "rewards/accuracies": 0.6499999761581421,
748
+ "rewards/chosen": -1.7339115142822266,
749
+ "rewards/margins": 0.8481589555740356,
750
+ "rewards/rejected": -2.5820705890655518,
751
+ "step": 520
752
+ },
753
+ {
754
+ "epoch": 0.42,
755
+ "learning_rate": 3.564448228912682e-06,
756
+ "logits/chosen": -1.6338014602661133,
757
+ "logits/rejected": -1.4595997333526611,
758
+ "logps/chosen": -781.9996948242188,
759
+ "logps/rejected": -803.5716552734375,
760
+ "loss": 0.6032,
761
+ "rewards/accuracies": 0.574999988079071,
762
+ "rewards/chosen": -1.587343454360962,
763
+ "rewards/margins": 0.3477286696434021,
764
+ "rewards/rejected": -1.9350721836090088,
765
+ "step": 530
766
+ },
767
+ {
768
+ "epoch": 0.43,
769
+ "learning_rate": 3.5008725813922383e-06,
770
+ "logits/chosen": -1.7276198863983154,
771
+ "logits/rejected": -1.5073800086975098,
772
+ "logps/chosen": -661.0886840820312,
773
+ "logps/rejected": -782.2366333007812,
774
+ "loss": 0.5138,
775
+ "rewards/accuracies": 0.7250000238418579,
776
+ "rewards/chosen": -0.5768153667449951,
777
+ "rewards/margins": 0.8154329061508179,
778
+ "rewards/rejected": -1.3922483921051025,
779
+ "step": 540
780
+ },
781
+ {
782
+ "epoch": 0.44,
783
+ "learning_rate": 3.436516483539781e-06,
784
+ "logits/chosen": -1.5888893604278564,
785
+ "logits/rejected": -1.5513622760772705,
786
+ "logps/chosen": -557.1135864257812,
787
+ "logps/rejected": -707.7156982421875,
788
+ "loss": 0.5318,
789
+ "rewards/accuracies": 0.574999988079071,
790
+ "rewards/chosen": -0.3991510272026062,
791
+ "rewards/margins": 0.7685856819152832,
792
+ "rewards/rejected": -1.1677367687225342,
793
+ "step": 550
794
+ },
795
+ {
796
+ "epoch": 0.45,
797
+ "learning_rate": 3.3714301183045382e-06,
798
+ "logits/chosen": -1.714202642440796,
799
+ "logits/rejected": -1.6166423559188843,
800
+ "logps/chosen": -575.7823486328125,
801
+ "logps/rejected": -702.8860473632812,
802
+ "loss": 0.555,
803
+ "rewards/accuracies": 0.6000000238418579,
804
+ "rewards/chosen": -0.6680876612663269,
805
+ "rewards/margins": 0.6836514472961426,
806
+ "rewards/rejected": -1.3517391681671143,
807
+ "step": 560
808
+ },
809
+ {
810
+ "epoch": 0.46,
811
+ "learning_rate": 3.3056642380762783e-06,
812
+ "logits/chosen": -1.7614761590957642,
813
+ "logits/rejected": -1.5657806396484375,
814
+ "logps/chosen": -635.3870849609375,
815
+ "logps/rejected": -756.2445678710938,
816
+ "loss": 0.5193,
817
+ "rewards/accuracies": 0.699999988079071,
818
+ "rewards/chosen": -0.9641979336738586,
819
+ "rewards/margins": 0.7494714856147766,
820
+ "rewards/rejected": -1.7136694192886353,
821
+ "step": 570
822
+ },
823
+ {
824
+ "epoch": 0.46,
825
+ "learning_rate": 3.2392701251101172e-06,
826
+ "logits/chosen": -1.9560362100601196,
827
+ "logits/rejected": -1.834413766860962,
828
+ "logps/chosen": -770.2546997070312,
829
+ "logps/rejected": -899.9256591796875,
830
+ "loss": 0.5241,
831
+ "rewards/accuracies": 0.625,
832
+ "rewards/chosen": -1.4754899740219116,
833
+ "rewards/margins": 0.7707339525222778,
834
+ "rewards/rejected": -2.2462239265441895,
835
+ "step": 580
836
+ },
837
+ {
838
+ "epoch": 0.47,
839
+ "learning_rate": 3.1722995515381644e-06,
840
+ "logits/chosen": -1.7270276546478271,
841
+ "logits/rejected": -1.4301753044128418,
842
+ "logps/chosen": -636.9214477539062,
843
+ "logps/rejected": -807.987548828125,
844
+ "loss": 0.5558,
845
+ "rewards/accuracies": 0.574999988079071,
846
+ "rewards/chosen": -1.3925626277923584,
847
+ "rewards/margins": 0.9666160345077515,
848
+ "rewards/rejected": -2.3591790199279785,
849
+ "step": 590
850
+ },
851
+ {
852
+ "epoch": 0.48,
853
+ "learning_rate": 3.1048047389991693e-06,
854
+ "logits/chosen": -1.5566425323486328,
855
+ "logits/rejected": -1.7140626907348633,
856
+ "logps/chosen": -586.4807739257812,
857
+ "logps/rejected": -803.2869873046875,
858
+ "loss": 0.5476,
859
+ "rewards/accuracies": 0.7250000238418579,
860
+ "rewards/chosen": -0.9375093579292297,
861
+ "rewards/margins": 1.2238131761550903,
862
+ "rewards/rejected": -2.1613223552703857,
863
+ "step": 600
864
+ },
865
+ {
866
+ "epoch": 0.49,
867
+ "learning_rate": 3.0368383179176584e-06,
868
+ "logits/chosen": -1.8013150691986084,
869
+ "logits/rejected": -1.8434633016586304,
870
+ "logps/chosen": -623.749267578125,
871
+ "logps/rejected": -685.603515625,
872
+ "loss": 0.5655,
873
+ "rewards/accuracies": 0.5249999761581421,
874
+ "rewards/chosen": -1.1918621063232422,
875
+ "rewards/margins": 0.3744085431098938,
876
+ "rewards/rejected": -1.5662705898284912,
877
+ "step": 610
878
+ },
879
+ {
880
+ "epoch": 0.5,
881
+ "learning_rate": 2.9684532864643123e-06,
882
+ "logits/chosen": -1.690509557723999,
883
+ "logits/rejected": -1.591670274734497,
884
+ "logps/chosen": -606.4476318359375,
885
+ "logps/rejected": -794.9050903320312,
886
+ "loss": 0.5668,
887
+ "rewards/accuracies": 0.699999988079071,
888
+ "rewards/chosen": -0.9053556323051453,
889
+ "rewards/margins": 0.9282801747322083,
890
+ "rewards/rejected": -1.833635687828064,
891
+ "step": 620
892
+ },
893
+ {
894
+ "epoch": 0.5,
895
+ "learning_rate": 2.8997029692295875e-06,
896
+ "logits/chosen": -1.6267430782318115,
897
+ "logits/rejected": -1.6104875802993774,
898
+ "logps/chosen": -682.794677734375,
899
+ "logps/rejected": -859.1971435546875,
900
+ "loss": 0.5286,
901
+ "rewards/accuracies": 0.699999988079071,
902
+ "rewards/chosen": -1.2506405115127563,
903
+ "rewards/margins": 0.5816906690597534,
904
+ "rewards/rejected": -1.8323312997817993,
905
+ "step": 630
906
+ },
907
+ {
908
+ "epoch": 0.51,
909
+ "learning_rate": 2.8306409756428067e-06,
910
+ "logits/chosen": -1.6168702840805054,
911
+ "logits/rejected": -1.510899305343628,
912
+ "logps/chosen": -718.5562133789062,
913
+ "logps/rejected": -864.6832275390625,
914
+ "loss": 0.5219,
915
+ "rewards/accuracies": 0.625,
916
+ "rewards/chosen": -1.8838809728622437,
917
+ "rewards/margins": 0.7105566263198853,
918
+ "rewards/rejected": -2.594438076019287,
919
+ "step": 640
920
+ },
921
+ {
922
+ "epoch": 0.52,
923
+ "learning_rate": 2.761321158169134e-06,
924
+ "logits/chosen": -1.7204669713974,
925
+ "logits/rejected": -1.5405805110931396,
926
+ "logps/chosen": -809.19482421875,
927
+ "logps/rejected": -887.3616943359375,
928
+ "loss": 0.521,
929
+ "rewards/accuracies": 0.574999988079071,
930
+ "rewards/chosen": -2.5423672199249268,
931
+ "rewards/margins": 0.6207980513572693,
932
+ "rewards/rejected": -3.163165330886841,
933
+ "step": 650
934
+ },
935
+ {
936
+ "epoch": 0.53,
937
+ "learning_rate": 2.6917975703170466e-06,
938
+ "logits/chosen": -1.5943357944488525,
939
+ "logits/rejected": -1.4551645517349243,
940
+ "logps/chosen": -739.6907348632812,
941
+ "logps/rejected": -910.01611328125,
942
+ "loss": 0.5359,
943
+ "rewards/accuracies": 0.625,
944
+ "rewards/chosen": -2.378048896789551,
945
+ "rewards/margins": 0.8076974749565125,
946
+ "rewards/rejected": -3.185746192932129,
947
+ "step": 660
948
+ },
949
+ {
950
+ "epoch": 0.54,
951
+ "learning_rate": 2.6221244244890336e-06,
952
+ "logits/chosen": -1.9027698040008545,
953
+ "logits/rejected": -1.643210768699646,
954
+ "logps/chosen": -800.6276245117188,
955
+ "logps/rejected": -961.8733520507812,
956
+ "loss": 0.5139,
957
+ "rewards/accuracies": 0.699999988079071,
958
+ "rewards/chosen": -1.9278568029403687,
959
+ "rewards/margins": 1.0248727798461914,
960
+ "rewards/rejected": -2.9527297019958496,
961
+ "step": 670
962
+ },
963
+ {
964
+ "epoch": 0.54,
965
+ "learning_rate": 2.5523560497083927e-06,
966
+ "logits/chosen": -1.6991322040557861,
967
+ "logits/rejected": -1.640962839126587,
968
+ "logps/chosen": -696.5242309570312,
969
+ "logps/rejected": -832.0681762695312,
970
+ "loss": 0.5859,
971
+ "rewards/accuracies": 0.574999988079071,
972
+ "rewards/chosen": -1.6130882501602173,
973
+ "rewards/margins": 0.5149592161178589,
974
+ "rewards/rejected": -2.128047466278076,
975
+ "step": 680
976
+ },
977
+ {
978
+ "epoch": 0.55,
979
+ "learning_rate": 2.482546849255096e-06,
980
+ "logits/chosen": -1.9012749195098877,
981
+ "logits/rejected": -1.6301822662353516,
982
+ "logps/chosen": -575.906982421875,
983
+ "logps/rejected": -705.2581787109375,
984
+ "loss": 0.5491,
985
+ "rewards/accuracies": 0.625,
986
+ "rewards/chosen": -1.1113747358322144,
987
+ "rewards/margins": 0.5878103971481323,
988
+ "rewards/rejected": -1.6991851329803467,
989
+ "step": 690
990
+ },
991
+ {
992
+ "epoch": 0.56,
993
+ "learning_rate": 2.4127512582437486e-06,
994
+ "logits/chosen": -1.9275786876678467,
995
+ "logits/rejected": -1.5215277671813965,
996
+ "logps/chosen": -685.0286865234375,
997
+ "logps/rejected": -747.36328125,
998
+ "loss": 0.5836,
999
+ "rewards/accuracies": 0.625,
1000
+ "rewards/chosen": -1.703952431678772,
1001
+ "rewards/margins": 0.682808518409729,
1002
+ "rewards/rejected": -2.386760711669922,
1003
+ "step": 700
1004
+ },
1005
+ {
1006
+ "epoch": 0.57,
1007
+ "learning_rate": 2.3430237011767166e-06,
1008
+ "logits/chosen": -1.877598524093628,
1009
+ "logits/rejected": -1.7826967239379883,
1010
+ "logps/chosen": -795.4154663085938,
1011
+ "logps/rejected": -927.1178588867188,
1012
+ "loss": 0.5641,
1013
+ "rewards/accuracies": 0.6499999761581421,
1014
+ "rewards/chosen": -2.108426570892334,
1015
+ "rewards/margins": 0.692725658416748,
1016
+ "rewards/rejected": -2.801152229309082,
1017
+ "step": 710
1018
+ },
1019
+ {
1020
+ "epoch": 0.58,
1021
+ "learning_rate": 2.2734185495055503e-06,
1022
+ "logits/chosen": -1.882228136062622,
1023
+ "logits/rejected": -1.8573133945465088,
1024
+ "logps/chosen": -863.8942260742188,
1025
+ "logps/rejected": -941.5260620117188,
1026
+ "loss": 0.5286,
1027
+ "rewards/accuracies": 0.574999988079071,
1028
+ "rewards/chosen": -2.1344828605651855,
1029
+ "rewards/margins": 0.45495691895484924,
1030
+ "rewards/rejected": -2.589439868927002,
1031
+ "step": 720
1032
+ },
1033
+ {
1034
+ "epoch": 0.58,
1035
+ "learning_rate": 2.2039900792337477e-06,
1036
+ "logits/chosen": -1.7438204288482666,
1037
+ "logits/rejected": -1.4300575256347656,
1038
+ "logps/chosen": -746.1878051757812,
1039
+ "logps/rejected": -917.93310546875,
1040
+ "loss": 0.5253,
1041
+ "rewards/accuracies": 0.675000011920929,
1042
+ "rewards/chosen": -1.661818504333496,
1043
+ "rewards/margins": 0.9328581094741821,
1044
+ "rewards/rejected": -2.5946764945983887,
1045
+ "step": 730
1046
+ },
1047
+ {
1048
+ "epoch": 0.59,
1049
+ "learning_rate": 2.134792428593971e-06,
1050
+ "logits/chosen": -1.9786335229873657,
1051
+ "logits/rejected": -1.641619324684143,
1052
+ "logps/chosen": -810.0553588867188,
1053
+ "logps/rejected": -976.5023193359375,
1054
+ "loss": 0.5209,
1055
+ "rewards/accuracies": 0.7250000238418579,
1056
+ "rewards/chosen": -1.9911285638809204,
1057
+ "rewards/margins": 0.9599992036819458,
1058
+ "rewards/rejected": -2.951127529144287,
1059
+ "step": 740
1060
+ },
1061
+ {
1062
+ "epoch": 0.6,
1063
+ "learning_rate": 2.0658795558326745e-06,
1064
+ "logits/chosen": -1.6420676708221436,
1065
+ "logits/rejected": -1.5285022258758545,
1066
+ "logps/chosen": -786.351806640625,
1067
+ "logps/rejected": -920.1495971679688,
1068
+ "loss": 0.5636,
1069
+ "rewards/accuracies": 0.6499999761581421,
1070
+ "rewards/chosen": -2.1379361152648926,
1071
+ "rewards/margins": 0.7854518294334412,
1072
+ "rewards/rejected": -2.9233880043029785,
1073
+ "step": 750
1074
+ },
1075
+ {
1076
+ "epoch": 0.61,
1077
+ "learning_rate": 1.997305197135089e-06,
1078
+ "logits/chosen": -1.6396596431732178,
1079
+ "logits/rejected": -1.707977533340454,
1080
+ "logps/chosen": -749.3070068359375,
1081
+ "logps/rejected": -999.5471801757812,
1082
+ "loss": 0.5326,
1083
+ "rewards/accuracies": 0.6499999761581421,
1084
+ "rewards/chosen": -2.0772664546966553,
1085
+ "rewards/margins": 0.8256868124008179,
1086
+ "rewards/rejected": -2.902953624725342,
1087
+ "step": 760
1088
+ },
1089
+ {
1090
+ "epoch": 0.62,
1091
+ "learning_rate": 1.9291228247233607e-06,
1092
+ "logits/chosen": -1.8173869848251343,
1093
+ "logits/rejected": -1.6075998544692993,
1094
+ "logps/chosen": -792.2544555664062,
1095
+ "logps/rejected": -778.5856323242188,
1096
+ "loss": 0.6253,
1097
+ "rewards/accuracies": 0.4000000059604645,
1098
+ "rewards/chosen": -2.354640245437622,
1099
+ "rewards/margins": 0.19878435134887695,
1100
+ "rewards/rejected": -2.55342435836792,
1101
+ "step": 770
1102
+ },
1103
+ {
1104
+ "epoch": 0.62,
1105
+ "learning_rate": 1.8613856051605242e-06,
1106
+ "logits/chosen": -1.9148937463760376,
1107
+ "logits/rejected": -1.708414077758789,
1108
+ "logps/chosen": -806.0308227539062,
1109
+ "logps/rejected": -816.46728515625,
1110
+ "loss": 0.5816,
1111
+ "rewards/accuracies": 0.5,
1112
+ "rewards/chosen": -2.1668009757995605,
1113
+ "rewards/margins": 0.1528003066778183,
1114
+ "rewards/rejected": -2.319601535797119,
1115
+ "step": 780
1116
+ },
1117
+ {
1118
+ "epoch": 0.63,
1119
+ "learning_rate": 1.7941463578928088e-06,
1120
+ "logits/chosen": -1.8512544631958008,
1121
+ "logits/rejected": -1.652984380722046,
1122
+ "logps/chosen": -732.0314331054688,
1123
+ "logps/rejected": -881.2513427734375,
1124
+ "loss": 0.5316,
1125
+ "rewards/accuracies": 0.699999988079071,
1126
+ "rewards/chosen": -1.7744696140289307,
1127
+ "rewards/margins": 0.6830210089683533,
1128
+ "rewards/rejected": -2.4574904441833496,
1129
+ "step": 790
1130
+ },
1131
+ {
1132
+ "epoch": 0.64,
1133
+ "learning_rate": 1.7274575140626318e-06,
1134
+ "logits/chosen": -1.8951454162597656,
1135
+ "logits/rejected": -1.6599276065826416,
1136
+ "logps/chosen": -631.180419921875,
1137
+ "logps/rejected": -686.8816528320312,
1138
+ "loss": 0.5847,
1139
+ "rewards/accuracies": 0.550000011920929,
1140
+ "rewards/chosen": -1.6252044439315796,
1141
+ "rewards/margins": 0.3886977434158325,
1142
+ "rewards/rejected": -2.013902187347412,
1143
+ "step": 800
1144
+ },
1145
+ {
1146
+ "epoch": 0.65,
1147
+ "learning_rate": 1.661371075624363e-06,
1148
+ "logits/chosen": -1.7395671606063843,
1149
+ "logits/rejected": -1.6132034063339233,
1150
+ "logps/chosen": -815.6490478515625,
1151
+ "logps/rejected": -1002.4617919921875,
1152
+ "loss": 0.5955,
1153
+ "rewards/accuracies": 0.625,
1154
+ "rewards/chosen": -2.1377925872802734,
1155
+ "rewards/margins": 1.0140737295150757,
1156
+ "rewards/rejected": -3.1518664360046387,
1157
+ "step": 810
1158
+ },
1159
+ {
1160
+ "epoch": 0.66,
1161
+ "learning_rate": 1.5959385747947697e-06,
1162
+ "logits/chosen": -1.823700189590454,
1163
+ "logits/rejected": -1.7485004663467407,
1164
+ "logps/chosen": -821.0921630859375,
1165
+ "logps/rejected": -1101.565185546875,
1166
+ "loss": 0.5003,
1167
+ "rewards/accuracies": 0.7749999761581421,
1168
+ "rewards/chosen": -2.3744442462921143,
1169
+ "rewards/margins": 1.0988423824310303,
1170
+ "rewards/rejected": -3.4732868671417236,
1171
+ "step": 820
1172
+ },
1173
+ {
1174
+ "epoch": 0.66,
1175
+ "learning_rate": 1.5312110338697427e-06,
1176
+ "logits/chosen": -1.6040973663330078,
1177
+ "logits/rejected": -1.4705755710601807,
1178
+ "logps/chosen": -791.6640625,
1179
+ "logps/rejected": -912.6233520507812,
1180
+ "loss": 0.5245,
1181
+ "rewards/accuracies": 0.625,
1182
+ "rewards/chosen": -2.226598024368286,
1183
+ "rewards/margins": 0.6341860294342041,
1184
+ "rewards/rejected": -2.8607842922210693,
1185
+ "step": 830
1186
+ },
1187
+ {
1188
+ "epoch": 0.67,
1189
+ "learning_rate": 1.467238925438646e-06,
1190
+ "logits/chosen": -1.7853944301605225,
1191
+ "logits/rejected": -1.6221034526824951,
1192
+ "logps/chosen": -775.5903930664062,
1193
+ "logps/rejected": -855.8067626953125,
1194
+ "loss": 0.585,
1195
+ "rewards/accuracies": 0.550000011920929,
1196
+ "rewards/chosen": -2.095076084136963,
1197
+ "rewards/margins": 0.3475552201271057,
1198
+ "rewards/rejected": -2.442631244659424,
1199
+ "step": 840
1200
+ },
1201
+ {
1202
+ "epoch": 0.68,
1203
+ "learning_rate": 1.4040721330273063e-06,
1204
+ "logits/chosen": -1.8012664318084717,
1205
+ "logits/rejected": -1.4391329288482666,
1206
+ "logps/chosen": -793.6468505859375,
1207
+ "logps/rejected": -884.8468627929688,
1208
+ "loss": 0.4651,
1209
+ "rewards/accuracies": 0.7250000238418579,
1210
+ "rewards/chosen": -2.0964674949645996,
1211
+ "rewards/margins": 0.7912064790725708,
1212
+ "rewards/rejected": -2.887673854827881,
1213
+ "step": 850
1214
+ },
1215
+ {
1216
+ "epoch": 0.69,
1217
+ "learning_rate": 1.3417599122003464e-06,
1218
+ "logits/chosen": -1.7759342193603516,
1219
+ "logits/rejected": -1.467376947402954,
1220
+ "logps/chosen": -779.6168823242188,
1221
+ "logps/rejected": -844.7642822265625,
1222
+ "loss": 0.5831,
1223
+ "rewards/accuracies": 0.550000011920929,
1224
+ "rewards/chosen": -2.3752453327178955,
1225
+ "rewards/margins": 0.3582616150379181,
1226
+ "rewards/rejected": -2.733506917953491,
1227
+ "step": 860
1228
+ },
1229
+ {
1230
+ "epoch": 0.7,
1231
+ "learning_rate": 1.280350852153168e-06,
1232
+ "logits/chosen": -1.8018977642059326,
1233
+ "logits/rejected": -1.644566297531128,
1234
+ "logps/chosen": -844.2316284179688,
1235
+ "logps/rejected": -997.5882568359375,
1236
+ "loss": 0.5303,
1237
+ "rewards/accuracies": 0.75,
1238
+ "rewards/chosen": -2.3328423500061035,
1239
+ "rewards/margins": 0.9274359941482544,
1240
+ "rewards/rejected": -3.2602787017822266,
1241
+ "step": 870
1242
+ },
1243
+ {
1244
+ "epoch": 0.7,
1245
+ "learning_rate": 1.2198928378235717e-06,
1246
+ "logits/chosen": -1.613319993019104,
1247
+ "logits/rejected": -1.5771634578704834,
1248
+ "logps/chosen": -804.1516723632812,
1249
+ "logps/rejected": -892.0836791992188,
1250
+ "loss": 0.5881,
1251
+ "rewards/accuracies": 0.574999988079071,
1252
+ "rewards/chosen": -2.6100995540618896,
1253
+ "rewards/margins": 0.4582017958164215,
1254
+ "rewards/rejected": -3.0683014392852783,
1255
+ "step": 880
1256
+ },
1257
+ {
1258
+ "epoch": 0.71,
1259
+ "learning_rate": 1.160433012552508e-06,
1260
+ "logits/chosen": -1.6768089532852173,
1261
+ "logits/rejected": -1.4050676822662354,
1262
+ "logps/chosen": -893.9998779296875,
1263
+ "logps/rejected": -987.3488159179688,
1264
+ "loss": 0.4851,
1265
+ "rewards/accuracies": 0.699999988079071,
1266
+ "rewards/chosen": -2.268981456756592,
1267
+ "rewards/margins": 0.9395302534103394,
1268
+ "rewards/rejected": -3.2085118293762207,
1269
+ "step": 890
1270
+ },
1271
+ {
1272
+ "epoch": 0.72,
1273
+ "learning_rate": 1.1020177413231334e-06,
1274
+ "logits/chosen": -1.8048509359359741,
1275
+ "logits/rejected": -1.6511070728302002,
1276
+ "logps/chosen": -758.7578125,
1277
+ "logps/rejected": -821.634765625,
1278
+ "loss": 0.5401,
1279
+ "rewards/accuracies": 0.550000011920929,
1280
+ "rewards/chosen": -2.2811436653137207,
1281
+ "rewards/margins": 0.47798076272010803,
1282
+ "rewards/rejected": -2.759124279022217,
1283
+ "step": 900
1284
+ },
1285
+ {
1286
+ "epoch": 0.73,
1287
+ "learning_rate": 1.0446925746067768e-06,
1288
+ "logits/chosen": -1.7981460094451904,
1289
+ "logits/rejected": -1.5802518129348755,
1290
+ "logps/chosen": -746.9505004882812,
1291
+ "logps/rejected": -945.9371337890625,
1292
+ "loss": 0.4969,
1293
+ "rewards/accuracies": 0.6000000238418579,
1294
+ "rewards/chosen": -2.0555319786071777,
1295
+ "rewards/margins": 0.9065355062484741,
1296
+ "rewards/rejected": -2.9620678424835205,
1297
+ "step": 910
1298
+ },
1299
+ {
1300
+ "epoch": 0.74,
1301
+ "learning_rate": 9.88502212844063e-07,
1302
+ "logits/chosen": -1.7775678634643555,
1303
+ "logits/rejected": -1.5710277557373047,
1304
+ "logps/chosen": -783.263671875,
1305
+ "logps/rejected": -912.9552001953125,
1306
+ "loss": 0.4923,
1307
+ "rewards/accuracies": 0.675000011920929,
1308
+ "rewards/chosen": -2.127746343612671,
1309
+ "rewards/margins": 0.6384310126304626,
1310
+ "rewards/rejected": -2.7661774158477783,
1311
+ "step": 920
1312
+ },
1313
+ {
1314
+ "epoch": 0.74,
1315
+ "learning_rate": 9.334904715888496e-07,
1316
+ "logits/chosen": -1.7053130865097046,
1317
+ "logits/rejected": -1.5473599433898926,
1318
+ "logps/chosen": -735.1968383789062,
1319
+ "logps/rejected": -902.2230224609375,
1320
+ "loss": 0.5515,
1321
+ "rewards/accuracies": 0.7250000238418579,
1322
+ "rewards/chosen": -1.9281718730926514,
1323
+ "rewards/margins": 0.9608237147331238,
1324
+ "rewards/rejected": -2.88899564743042,
1325
+ "step": 930
1326
+ },
1327
+ {
1328
+ "epoch": 0.75,
1329
+ "learning_rate": 8.797002473421729e-07,
1330
+ "logits/chosen": -1.7851415872573853,
1331
+ "logits/rejected": -1.6908962726593018,
1332
+ "logps/chosen": -669.7711181640625,
1333
+ "logps/rejected": -801.768310546875,
1334
+ "loss": 0.5386,
1335
+ "rewards/accuracies": 0.6000000238418579,
1336
+ "rewards/chosen": -1.95120108127594,
1337
+ "rewards/margins": 0.651104748249054,
1338
+ "rewards/rejected": -2.6023061275482178,
1339
+ "step": 940
1340
+ },
1341
+ {
1342
+ "epoch": 0.76,
1343
+ "learning_rate": 8.271734841028553e-07,
1344
+ "logits/chosen": -1.8082669973373413,
1345
+ "logits/rejected": -1.721047043800354,
1346
+ "logps/chosen": -722.6148681640625,
1347
+ "logps/rejected": -852.7736206054688,
1348
+ "loss": 0.5256,
1349
+ "rewards/accuracies": 0.5,
1350
+ "rewards/chosen": -2.0907461643218994,
1351
+ "rewards/margins": 0.7468978762626648,
1352
+ "rewards/rejected": -2.83764386177063,
1353
+ "step": 950
1354
+ },
1355
+ {
1356
+ "epoch": 0.77,
1357
+ "learning_rate": 7.759511406608255e-07,
1358
+ "logits/chosen": -1.7400856018066406,
1359
+ "logits/rejected": -1.3403172492980957,
1360
+ "logps/chosen": -768.3906860351562,
1361
+ "logps/rejected": -898.74462890625,
1362
+ "loss": 0.5437,
1363
+ "rewards/accuracies": 0.7250000238418579,
1364
+ "rewards/chosen": -1.9469468593597412,
1365
+ "rewards/margins": 1.0306456089019775,
1366
+ "rewards/rejected": -2.9775922298431396,
1367
+ "step": 960
1368
+ },
1369
+ {
1370
+ "epoch": 0.78,
1371
+ "learning_rate": 7.260731586586983e-07,
1372
+ "logits/chosen": -1.7103416919708252,
1373
+ "logits/rejected": -1.6440057754516602,
1374
+ "logps/chosen": -787.998046875,
1375
+ "logps/rejected": -901.8524169921875,
1376
+ "loss": 0.5215,
1377
+ "rewards/accuracies": 0.6499999761581421,
1378
+ "rewards/chosen": -2.1284966468811035,
1379
+ "rewards/margins": 0.7515214681625366,
1380
+ "rewards/rejected": -2.8800179958343506,
1381
+ "step": 970
1382
+ },
1383
+ {
1384
+ "epoch": 0.78,
1385
+ "learning_rate": 6.775784314464717e-07,
1386
+ "logits/chosen": -1.6210864782333374,
1387
+ "logits/rejected": -1.7291233539581299,
1388
+ "logps/chosen": -799.4705200195312,
1389
+ "logps/rejected": -899.9567260742188,
1390
+ "loss": 0.5348,
1391
+ "rewards/accuracies": 0.5249999761581421,
1392
+ "rewards/chosen": -2.247741460800171,
1393
+ "rewards/margins": 0.48886141180992126,
1394
+ "rewards/rejected": -2.736603021621704,
1395
+ "step": 980
1396
+ },
1397
+ {
1398
+ "epoch": 0.79,
1399
+ "learning_rate": 6.305047737536707e-07,
1400
+ "logits/chosen": -1.7531540393829346,
1401
+ "logits/rejected": -1.4522241353988647,
1402
+ "logps/chosen": -673.4083251953125,
1403
+ "logps/rejected": -805.2154541015625,
1404
+ "loss": 0.4932,
1405
+ "rewards/accuracies": 0.550000011920929,
1406
+ "rewards/chosen": -1.5855463743209839,
1407
+ "rewards/margins": 0.7592282891273499,
1408
+ "rewards/rejected": -2.3447744846343994,
1409
+ "step": 990
1410
+ },
1411
+ {
1412
+ "epoch": 0.8,
1413
+ "learning_rate": 5.848888922025553e-07,
1414
+ "logits/chosen": -1.8826286792755127,
1415
+ "logits/rejected": -1.697880744934082,
1416
+ "logps/chosen": -639.0965576171875,
1417
+ "logps/rejected": -673.7703857421875,
1418
+ "loss": 0.5212,
1419
+ "rewards/accuracies": 0.6000000238418579,
1420
+ "rewards/chosen": -1.3480440378189087,
1421
+ "rewards/margins": 0.4533866345882416,
1422
+ "rewards/rejected": -1.8014309406280518,
1423
+ "step": 1000
1424
+ },
1425
+ {
1426
+ "epoch": 0.81,
1427
+ "learning_rate": 5.407663566854008e-07,
1428
+ "logits/chosen": -1.715491533279419,
1429
+ "logits/rejected": -1.6110996007919312,
1430
+ "logps/chosen": -659.099609375,
1431
+ "logps/rejected": -858.7601318359375,
1432
+ "loss": 0.5271,
1433
+ "rewards/accuracies": 0.6499999761581421,
1434
+ "rewards/chosen": -1.3519906997680664,
1435
+ "rewards/margins": 0.9606063961982727,
1436
+ "rewards/rejected": -2.3125970363616943,
1437
+ "step": 1010
1438
+ },
1439
+ {
1440
+ "epoch": 0.82,
1441
+ "learning_rate": 4.981715726281666e-07,
1442
+ "logits/chosen": -1.8784500360488892,
1443
+ "logits/rejected": -1.659430742263794,
1444
+ "logps/chosen": -726.8290405273438,
1445
+ "logps/rejected": -912.7119140625,
1446
+ "loss": 0.5367,
1447
+ "rewards/accuracies": 0.699999988079071,
1448
+ "rewards/chosen": -1.4866254329681396,
1449
+ "rewards/margins": 1.1752581596374512,
1450
+ "rewards/rejected": -2.66188383102417,
1451
+ "step": 1020
1452
+ },
1453
+ {
1454
+ "epoch": 0.82,
1455
+ "learning_rate": 4.5713775416217884e-07,
1456
+ "logits/chosen": -1.9123560190200806,
1457
+ "logits/rejected": -1.732966423034668,
1458
+ "logps/chosen": -727.9432373046875,
1459
+ "logps/rejected": -894.9241333007812,
1460
+ "loss": 0.5749,
1461
+ "rewards/accuracies": 0.675000011920929,
1462
+ "rewards/chosen": -1.673956274986267,
1463
+ "rewards/margins": 0.8849543333053589,
1464
+ "rewards/rejected": -2.558910608291626,
1465
+ "step": 1030
1466
+ },
1467
+ {
1468
+ "epoch": 0.83,
1469
+ "learning_rate": 4.1769689822475147e-07,
1470
+ "logits/chosen": -1.8131048679351807,
1471
+ "logits/rejected": -1.6417629718780518,
1472
+ "logps/chosen": -647.2805786132812,
1473
+ "logps/rejected": -839.04345703125,
1474
+ "loss": 0.5169,
1475
+ "rewards/accuracies": 0.7250000238418579,
1476
+ "rewards/chosen": -1.5158132314682007,
1477
+ "rewards/margins": 0.8997069597244263,
1478
+ "rewards/rejected": -2.415520191192627,
1479
+ "step": 1040
1480
+ },
1481
+ {
1482
+ "epoch": 0.84,
1483
+ "learning_rate": 3.798797596089351e-07,
1484
+ "logits/chosen": -1.7820684909820557,
1485
+ "logits/rejected": -1.5671303272247314,
1486
+ "logps/chosen": -824.466796875,
1487
+ "logps/rejected": -998.0120239257812,
1488
+ "loss": 0.5417,
1489
+ "rewards/accuracies": 0.6000000238418579,
1490
+ "rewards/chosen": -2.170335054397583,
1491
+ "rewards/margins": 0.8759534955024719,
1492
+ "rewards/rejected": -3.046288251876831,
1493
+ "step": 1050
1494
+ },
1495
+ {
1496
+ "epoch": 0.85,
1497
+ "learning_rate": 3.4371582698185636e-07,
1498
+ "logits/chosen": -1.7961889505386353,
1499
+ "logits/rejected": -1.556156873703003,
1500
+ "logps/chosen": -742.6277465820312,
1501
+ "logps/rejected": -868.2706909179688,
1502
+ "loss": 0.5244,
1503
+ "rewards/accuracies": 0.625,
1504
+ "rewards/chosen": -1.7436439990997314,
1505
+ "rewards/margins": 0.8180065155029297,
1506
+ "rewards/rejected": -2.561650514602661,
1507
+ "step": 1060
1508
+ },
1509
+ {
1510
+ "epoch": 0.86,
1511
+ "learning_rate": 3.092332998903416e-07,
1512
+ "logits/chosen": -1.738315224647522,
1513
+ "logits/rejected": -1.4983936548233032,
1514
+ "logps/chosen": -744.6055297851562,
1515
+ "logps/rejected": -826.5979614257812,
1516
+ "loss": 0.5492,
1517
+ "rewards/accuracies": 0.6000000238418579,
1518
+ "rewards/chosen": -1.748814582824707,
1519
+ "rewards/margins": 0.5608989596366882,
1520
+ "rewards/rejected": -2.309713840484619,
1521
+ "step": 1070
1522
+ },
1523
+ {
1524
+ "epoch": 0.86,
1525
+ "learning_rate": 2.764590667717562e-07,
1526
+ "logits/chosen": -1.8366153240203857,
1527
+ "logits/rejected": -1.6621793508529663,
1528
+ "logps/chosen": -766.6265258789062,
1529
+ "logps/rejected": -869.9232177734375,
1530
+ "loss": 0.4771,
1531
+ "rewards/accuracies": 0.6000000238418579,
1532
+ "rewards/chosen": -1.6844078302383423,
1533
+ "rewards/margins": 0.7456134557723999,
1534
+ "rewards/rejected": -2.4300215244293213,
1535
+ "step": 1080
1536
+ },
1537
+ {
1538
+ "epoch": 0.87,
1539
+ "learning_rate": 2.454186839872158e-07,
1540
+ "logits/chosen": -1.752484917640686,
1541
+ "logits/rejected": -1.7045217752456665,
1542
+ "logps/chosen": -753.0501708984375,
1543
+ "logps/rejected": -849.8084716796875,
1544
+ "loss": 0.5826,
1545
+ "rewards/accuracies": 0.6499999761581421,
1546
+ "rewards/chosen": -1.7086284160614014,
1547
+ "rewards/margins": 0.6786292195320129,
1548
+ "rewards/rejected": -2.3872575759887695,
1549
+ "step": 1090
1550
+ },
1551
+ {
1552
+ "epoch": 0.88,
1553
+ "learning_rate": 2.1613635589349756e-07,
1554
+ "logits/chosen": -1.7320369482040405,
1555
+ "logits/rejected": -1.7084888219833374,
1556
+ "logps/chosen": -663.6228637695312,
1557
+ "logps/rejected": -909.4744262695312,
1558
+ "loss": 0.5444,
1559
+ "rewards/accuracies": 0.7749999761581421,
1560
+ "rewards/chosen": -1.5473442077636719,
1561
+ "rewards/margins": 1.0618083477020264,
1562
+ "rewards/rejected": -2.6091525554656982,
1563
+ "step": 1100
1564
+ },
1565
+ {
1566
+ "epoch": 0.89,
1567
+ "learning_rate": 1.8863491596921745e-07,
1568
+ "logits/chosen": -1.8900954723358154,
1569
+ "logits/rejected": -1.5332658290863037,
1570
+ "logps/chosen": -759.9290161132812,
1571
+ "logps/rejected": -967.7333984375,
1572
+ "loss": 0.4666,
1573
+ "rewards/accuracies": 0.800000011920929,
1574
+ "rewards/chosen": -1.6536929607391357,
1575
+ "rewards/margins": 1.2927501201629639,
1576
+ "rewards/rejected": -2.9464430809020996,
1577
+ "step": 1110
1578
+ },
1579
+ {
1580
+ "epoch": 0.9,
1581
+ "learning_rate": 1.629358090099639e-07,
1582
+ "logits/chosen": -1.7427972555160522,
1583
+ "logits/rejected": -1.7908833026885986,
1584
+ "logps/chosen": -699.6878662109375,
1585
+ "logps/rejected": -807.19140625,
1586
+ "loss": 0.5983,
1587
+ "rewards/accuracies": 0.5,
1588
+ "rewards/chosen": -1.8040729761123657,
1589
+ "rewards/margins": 0.4588094651699066,
1590
+ "rewards/rejected": -2.262882709503174,
1591
+ "step": 1120
1592
+ },
1593
+ {
1594
+ "epoch": 0.9,
1595
+ "learning_rate": 1.3905907440629752e-07,
1596
+ "logits/chosen": -1.7846444845199585,
1597
+ "logits/rejected": -1.638474702835083,
1598
+ "logps/chosen": -695.3917236328125,
1599
+ "logps/rejected": -831.669921875,
1600
+ "loss": 0.5498,
1601
+ "rewards/accuracies": 0.5249999761581421,
1602
+ "rewards/chosen": -1.7915891408920288,
1603
+ "rewards/margins": 0.6667860150337219,
1604
+ "rewards/rejected": -2.4583749771118164,
1605
+ "step": 1130
1606
+ },
1607
+ {
1608
+ "epoch": 0.91,
1609
+ "learning_rate": 1.1702333051763271e-07,
1610
+ "logits/chosen": -1.9200313091278076,
1611
+ "logits/rejected": -1.6578702926635742,
1612
+ "logps/chosen": -790.5162353515625,
1613
+ "logps/rejected": -900.2391357421875,
1614
+ "loss": 0.5435,
1615
+ "rewards/accuracies": 0.550000011920929,
1616
+ "rewards/chosen": -1.8958162069320679,
1617
+ "rewards/margins": 0.7216595411300659,
1618
+ "rewards/rejected": -2.6174755096435547,
1619
+ "step": 1140
1620
+ },
1621
+ {
1622
+ "epoch": 0.92,
1623
+ "learning_rate": 9.684576015420277e-08,
1624
+ "logits/chosen": -1.9687907695770264,
1625
+ "logits/rejected": -1.5580371618270874,
1626
+ "logps/chosen": -751.2023315429688,
1627
+ "logps/rejected": -922.86865234375,
1628
+ "loss": 0.4658,
1629
+ "rewards/accuracies": 0.75,
1630
+ "rewards/chosen": -1.6761118173599243,
1631
+ "rewards/margins": 1.1890485286712646,
1632
+ "rewards/rejected": -2.8651604652404785,
1633
+ "step": 1150
1634
+ },
1635
+ {
1636
+ "epoch": 0.93,
1637
+ "learning_rate": 7.854209717842231e-08,
1638
+ "logits/chosen": -1.7421581745147705,
1639
+ "logits/rejected": -1.599867582321167,
1640
+ "logps/chosen": -746.3426513671875,
1641
+ "logps/rejected": -906.96044921875,
1642
+ "loss": 0.5019,
1643
+ "rewards/accuracies": 0.574999988079071,
1644
+ "rewards/chosen": -1.9812192916870117,
1645
+ "rewards/margins": 0.9240058660507202,
1646
+ "rewards/rejected": -2.9052250385284424,
1647
+ "step": 1160
1648
+ },
1649
+ {
1650
+ "epoch": 0.94,
1651
+ "learning_rate": 6.212661423609184e-08,
1652
+ "logits/chosen": -1.8295780420303345,
1653
+ "logits/rejected": -1.6900854110717773,
1654
+ "logps/chosen": -664.9674072265625,
1655
+ "logps/rejected": -825.6485595703125,
1656
+ "loss": 0.4682,
1657
+ "rewards/accuracies": 0.574999988079071,
1658
+ "rewards/chosen": -1.7017185688018799,
1659
+ "rewards/margins": 0.8063501119613647,
1660
+ "rewards/rejected": -2.5080690383911133,
1661
+ "step": 1170
1662
+ },
1663
+ {
1664
+ "epoch": 0.94,
1665
+ "learning_rate": 4.761211162702117e-08,
1666
+ "logits/chosen": -1.7115318775177002,
1667
+ "logits/rejected": -1.5249701738357544,
1668
+ "logps/chosen": -784.123291015625,
1669
+ "logps/rejected": -991.6115112304688,
1670
+ "loss": 0.5539,
1671
+ "rewards/accuracies": 0.6499999761581421,
1672
+ "rewards/chosen": -2.106039047241211,
1673
+ "rewards/margins": 0.8404006958007812,
1674
+ "rewards/rejected": -2.946439743041992,
1675
+ "step": 1180
1676
+ },
1677
+ {
1678
+ "epoch": 0.95,
1679
+ "learning_rate": 3.5009907323737826e-08,
1680
+ "logits/chosen": -1.8599879741668701,
1681
+ "logits/rejected": -1.5368030071258545,
1682
+ "logps/chosen": -662.7159423828125,
1683
+ "logps/rejected": -818.0572509765625,
1684
+ "loss": 0.4828,
1685
+ "rewards/accuracies": 0.6499999761581421,
1686
+ "rewards/chosen": -1.505114197731018,
1687
+ "rewards/margins": 0.9477652311325073,
1688
+ "rewards/rejected": -2.4528794288635254,
1689
+ "step": 1190
1690
+ },
1691
+ {
1692
+ "epoch": 0.96,
1693
+ "learning_rate": 2.4329828146074096e-08,
1694
+ "logits/chosen": -1.753741979598999,
1695
+ "logits/rejected": -1.6953494548797607,
1696
+ "logps/chosen": -697.6729736328125,
1697
+ "logps/rejected": -896.8612060546875,
1698
+ "loss": 0.5336,
1699
+ "rewards/accuracies": 0.7250000238418579,
1700
+ "rewards/chosen": -1.672422170639038,
1701
+ "rewards/margins": 0.98626708984375,
1702
+ "rewards/rejected": -2.658689260482788,
1703
+ "step": 1200
1704
+ },
1705
+ {
1706
+ "epoch": 0.97,
1707
+ "learning_rate": 1.5580202098509078e-08,
1708
+ "logits/chosen": -1.7109054327011108,
1709
+ "logits/rejected": -1.6440505981445312,
1710
+ "logps/chosen": -711.5040283203125,
1711
+ "logps/rejected": -841.7639770507812,
1712
+ "loss": 0.5112,
1713
+ "rewards/accuracies": 0.625,
1714
+ "rewards/chosen": -1.7976610660552979,
1715
+ "rewards/margins": 0.7495275735855103,
1716
+ "rewards/rejected": -2.5471885204315186,
1717
+ "step": 1210
1718
+ },
1719
+ {
1720
+ "epoch": 0.98,
1721
+ "learning_rate": 8.767851876239075e-09,
1722
+ "logits/chosen": -1.7670787572860718,
1723
+ "logits/rejected": -1.4884636402130127,
1724
+ "logps/chosen": -749.7885131835938,
1725
+ "logps/rejected": -868.7528076171875,
1726
+ "loss": 0.5099,
1727
+ "rewards/accuracies": 0.574999988079071,
1728
+ "rewards/chosen": -2.0942044258117676,
1729
+ "rewards/margins": 0.6244227290153503,
1730
+ "rewards/rejected": -2.718627452850342,
1731
+ "step": 1220
1732
+ },
1733
+ {
1734
+ "epoch": 0.98,
1735
+ "learning_rate": 3.8980895450474455e-09,
1736
+ "logits/chosen": -1.7517249584197998,
1737
+ "logits/rejected": -1.351718544960022,
1738
+ "logps/chosen": -748.0111083984375,
1739
+ "logps/rejected": -896.0007934570312,
1740
+ "loss": 0.4875,
1741
+ "rewards/accuracies": 0.7250000238418579,
1742
+ "rewards/chosen": -1.6639865636825562,
1743
+ "rewards/margins": 1.0765060186386108,
1744
+ "rewards/rejected": -2.740492820739746,
1745
+ "step": 1230
1746
+ },
1747
+ {
1748
+ "epoch": 0.99,
1749
+ "learning_rate": 9.747123991141193e-10,
1750
+ "logits/chosen": -1.8468574285507202,
1751
+ "logits/rejected": -1.5728193521499634,
1752
+ "logps/chosen": -750.45947265625,
1753
+ "logps/rejected": -858.3756713867188,
1754
+ "loss": 0.5676,
1755
+ "rewards/accuracies": 0.75,
1756
+ "rewards/chosen": -1.781569480895996,
1757
+ "rewards/margins": 0.7082441449165344,
1758
+ "rewards/rejected": -2.489814043045044,
1759
+ "step": 1240
1760
+ },
1761
+ {
1762
+ "epoch": 1.0,
1763
+ "learning_rate": 0.0,
1764
+ "logits/chosen": -1.9796960353851318,
1765
+ "logits/rejected": -1.5188578367233276,
1766
+ "logps/chosen": -744.5242309570312,
1767
+ "logps/rejected": -853.5891723632812,
1768
+ "loss": 0.5216,
1769
+ "rewards/accuracies": 0.675000011920929,
1770
+ "rewards/chosen": -1.9095113277435303,
1771
+ "rewards/margins": 0.9789590835571289,
1772
+ "rewards/rejected": -2.8884706497192383,
1773
+ "step": 1250
1774
+ },
1775
+ {
1776
+ "epoch": 1.0,
1777
+ "step": 1250,
1778
+ "total_flos": 0.0,
1779
+ "train_loss": 0.5400541301727295,
1780
+ "train_runtime": 12749.1866,
1781
+ "train_samples_per_second": 1.177,
1782
+ "train_steps_per_second": 0.098
1783
+ }
1784
+ ],
1785
+ "logging_steps": 10,
1786
+ "max_steps": 1250,
1787
+ "num_input_tokens_seen": 0,
1788
+ "num_train_epochs": 1,
1789
+ "save_steps": 20,
1790
+ "total_flos": 0.0,
1791
+ "train_batch_size": 2,
1792
+ "trial_name": null,
1793
+ "trial_params": null
1794
+ }