Model save
Browse files- README.md +76 -0
- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +356 -0
README.md
ADDED
|
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- dpo
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
base_model: alignment-handbook/zephyr-7b-sft-full
|
| 9 |
+
model-index:
|
| 10 |
+
- name: nash_dpo_merge_iter_6
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# nash_dpo_merge_iter_6
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on the None dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.5366
|
| 22 |
+
- Rewards/chosen: -0.8030
|
| 23 |
+
- Rewards/rejected: -1.4875
|
| 24 |
+
- Rewards/accuracies: 0.7320
|
| 25 |
+
- Rewards/margins: 0.6845
|
| 26 |
+
- Logps/rejected: -415.7152
|
| 27 |
+
- Logps/chosen: -367.6444
|
| 28 |
+
- Logits/rejected: -0.6068
|
| 29 |
+
- Logits/chosen: -1.1801
|
| 30 |
+
|
| 31 |
+
## Model description
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Intended uses & limitations
|
| 36 |
+
|
| 37 |
+
More information needed
|
| 38 |
+
|
| 39 |
+
## Training and evaluation data
|
| 40 |
+
|
| 41 |
+
More information needed
|
| 42 |
+
|
| 43 |
+
## Training procedure
|
| 44 |
+
|
| 45 |
+
### Training hyperparameters
|
| 46 |
+
|
| 47 |
+
The following hyperparameters were used during training:
|
| 48 |
+
- learning_rate: 5e-06
|
| 49 |
+
- train_batch_size: 2
|
| 50 |
+
- eval_batch_size: 2
|
| 51 |
+
- seed: 42
|
| 52 |
+
- distributed_type: multi-GPU
|
| 53 |
+
- num_devices: 4
|
| 54 |
+
- gradient_accumulation_steps: 16
|
| 55 |
+
- total_train_batch_size: 128
|
| 56 |
+
- total_eval_batch_size: 8
|
| 57 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 58 |
+
- lr_scheduler_type: cosine
|
| 59 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 60 |
+
- num_epochs: 1
|
| 61 |
+
|
| 62 |
+
### Training results
|
| 63 |
+
|
| 64 |
+
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|
| 65 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
|
| 66 |
+
| 0.515 | 0.49 | 100 | 0.5449 | -0.7799 | -1.4614 | 0.7160 | 0.6816 | -413.1093 | -365.3295 | -0.5798 | -1.2092 |
|
| 67 |
+
| 0.4925 | 0.98 | 200 | 0.5366 | -0.8030 | -1.4875 | 0.7320 | 0.6845 | -415.7152 | -367.6444 | -0.6068 | -1.1801 |
|
| 68 |
+
|
| 69 |
+
|
| 70 |
+
### Framework versions
|
| 71 |
+
|
| 72 |
+
- PEFT 0.7.1
|
| 73 |
+
- Transformers 4.36.2
|
| 74 |
+
- Pytorch 2.1.2+cu121
|
| 75 |
+
- Datasets 2.14.6
|
| 76 |
+
- Tokenizers 0.15.2
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.54004861034599,
|
| 4 |
+
"train_runtime": 9618.5725,
|
| 5 |
+
"train_samples": 26135,
|
| 6 |
+
"train_samples_per_second": 2.717,
|
| 7 |
+
"train_steps_per_second": 0.021
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.54004861034599,
|
| 4 |
+
"train_runtime": 9618.5725,
|
| 5 |
+
"train_samples": 26135,
|
| 6 |
+
"train_samples_per_second": 2.717,
|
| 7 |
+
"train_steps_per_second": 0.021
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,356 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9990817263544536,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 204,
|
| 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": 2.3809523809523811e-07,
|
| 14 |
+
"logits/chosen": -2.476858377456665,
|
| 15 |
+
"logits/rejected": -2.312589406967163,
|
| 16 |
+
"logps/chosen": -256.081787109375,
|
| 17 |
+
"logps/rejected": -237.47195434570312,
|
| 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": 0.05,
|
| 27 |
+
"learning_rate": 2.380952380952381e-06,
|
| 28 |
+
"logits/chosen": -2.2722971439361572,
|
| 29 |
+
"logits/rejected": -2.247359275817871,
|
| 30 |
+
"logps/chosen": -218.2034454345703,
|
| 31 |
+
"logps/rejected": -239.2631072998047,
|
| 32 |
+
"loss": 0.6892,
|
| 33 |
+
"rewards/accuracies": 0.5277777910232544,
|
| 34 |
+
"rewards/chosen": 0.0009012501686811447,
|
| 35 |
+
"rewards/margins": 0.007593905553221703,
|
| 36 |
+
"rewards/rejected": -0.0066926549188792706,
|
| 37 |
+
"step": 10
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.1,
|
| 41 |
+
"learning_rate": 4.761904761904762e-06,
|
| 42 |
+
"logits/chosen": -1.9851748943328857,
|
| 43 |
+
"logits/rejected": -1.7117283344268799,
|
| 44 |
+
"logps/chosen": -327.2151794433594,
|
| 45 |
+
"logps/rejected": -315.92840576171875,
|
| 46 |
+
"loss": 0.6161,
|
| 47 |
+
"rewards/accuracies": 0.71875,
|
| 48 |
+
"rewards/chosen": -0.29775306582450867,
|
| 49 |
+
"rewards/margins": 0.21288084983825684,
|
| 50 |
+
"rewards/rejected": -0.5106339454650879,
|
| 51 |
+
"step": 20
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.15,
|
| 55 |
+
"learning_rate": 4.970219740227693e-06,
|
| 56 |
+
"logits/chosen": -1.6041641235351562,
|
| 57 |
+
"logits/rejected": -1.3207042217254639,
|
| 58 |
+
"logps/chosen": -298.0397644042969,
|
| 59 |
+
"logps/rejected": -315.13543701171875,
|
| 60 |
+
"loss": 0.5858,
|
| 61 |
+
"rewards/accuracies": 0.703125,
|
| 62 |
+
"rewards/chosen": -0.27262887358665466,
|
| 63 |
+
"rewards/margins": 0.3364975154399872,
|
| 64 |
+
"rewards/rejected": -0.6091263294219971,
|
| 65 |
+
"step": 30
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.2,
|
| 69 |
+
"learning_rate": 4.868186180746792e-06,
|
| 70 |
+
"logits/chosen": -1.2411524057388306,
|
| 71 |
+
"logits/rejected": -0.8121637105941772,
|
| 72 |
+
"logps/chosen": -354.04315185546875,
|
| 73 |
+
"logps/rejected": -402.13787841796875,
|
| 74 |
+
"loss": 0.5475,
|
| 75 |
+
"rewards/accuracies": 0.7250000238418579,
|
| 76 |
+
"rewards/chosen": -0.5969960689544678,
|
| 77 |
+
"rewards/margins": 0.565431535243988,
|
| 78 |
+
"rewards/rejected": -1.1624276638031006,
|
| 79 |
+
"step": 40
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.24,
|
| 83 |
+
"learning_rate": 4.696530612642871e-06,
|
| 84 |
+
"logits/chosen": -1.148498296737671,
|
| 85 |
+
"logits/rejected": -0.7622831463813782,
|
| 86 |
+
"logps/chosen": -342.1610107421875,
|
| 87 |
+
"logps/rejected": -401.2055969238281,
|
| 88 |
+
"loss": 0.5384,
|
| 89 |
+
"rewards/accuracies": 0.7250000238418579,
|
| 90 |
+
"rewards/chosen": -0.5624964833259583,
|
| 91 |
+
"rewards/margins": 0.587742030620575,
|
| 92 |
+
"rewards/rejected": -1.1502383947372437,
|
| 93 |
+
"step": 50
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.29,
|
| 97 |
+
"learning_rate": 4.460299516441777e-06,
|
| 98 |
+
"logits/chosen": -1.0859754085540771,
|
| 99 |
+
"logits/rejected": -0.6681124567985535,
|
| 100 |
+
"logps/chosen": -328.865234375,
|
| 101 |
+
"logps/rejected": -390.8894348144531,
|
| 102 |
+
"loss": 0.5481,
|
| 103 |
+
"rewards/accuracies": 0.734375,
|
| 104 |
+
"rewards/chosen": -0.5670114159584045,
|
| 105 |
+
"rewards/margins": 0.5836409330368042,
|
| 106 |
+
"rewards/rejected": -1.150652289390564,
|
| 107 |
+
"step": 60
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.34,
|
| 111 |
+
"learning_rate": 4.1664378205239085e-06,
|
| 112 |
+
"logits/chosen": -0.855512797832489,
|
| 113 |
+
"logits/rejected": -0.3595692217350006,
|
| 114 |
+
"logps/chosen": -353.48675537109375,
|
| 115 |
+
"logps/rejected": -402.8636779785156,
|
| 116 |
+
"loss": 0.5327,
|
| 117 |
+
"rewards/accuracies": 0.699999988079071,
|
| 118 |
+
"rewards/chosen": -0.8263834714889526,
|
| 119 |
+
"rewards/margins": 0.5548129081726074,
|
| 120 |
+
"rewards/rejected": -1.38119637966156,
|
| 121 |
+
"step": 70
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.39,
|
| 125 |
+
"learning_rate": 3.8235847280454626e-06,
|
| 126 |
+
"logits/chosen": -1.166483998298645,
|
| 127 |
+
"logits/rejected": -0.6420685052871704,
|
| 128 |
+
"logps/chosen": -332.43975830078125,
|
| 129 |
+
"logps/rejected": -391.0145263671875,
|
| 130 |
+
"loss": 0.4927,
|
| 131 |
+
"rewards/accuracies": 0.753125011920929,
|
| 132 |
+
"rewards/chosen": -0.5521113276481628,
|
| 133 |
+
"rewards/margins": 0.6991725564002991,
|
| 134 |
+
"rewards/rejected": -1.251283884048462,
|
| 135 |
+
"step": 80
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.44,
|
| 139 |
+
"learning_rate": 3.441819734087963e-06,
|
| 140 |
+
"logits/chosen": -1.4997479915618896,
|
| 141 |
+
"logits/rejected": -0.9499446153640747,
|
| 142 |
+
"logps/chosen": -348.1226806640625,
|
| 143 |
+
"logps/rejected": -402.1637268066406,
|
| 144 |
+
"loss": 0.5287,
|
| 145 |
+
"rewards/accuracies": 0.7437499761581421,
|
| 146 |
+
"rewards/chosen": -0.5159438848495483,
|
| 147 |
+
"rewards/margins": 0.7751549482345581,
|
| 148 |
+
"rewards/rejected": -1.2910988330841064,
|
| 149 |
+
"step": 90
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.49,
|
| 153 |
+
"learning_rate": 3.0323662998460396e-06,
|
| 154 |
+
"logits/chosen": -1.3590983152389526,
|
| 155 |
+
"logits/rejected": -0.7285458445549011,
|
| 156 |
+
"logps/chosen": -362.32666015625,
|
| 157 |
+
"logps/rejected": -408.63885498046875,
|
| 158 |
+
"loss": 0.515,
|
| 159 |
+
"rewards/accuracies": 0.706250011920929,
|
| 160 |
+
"rewards/chosen": -0.7177928686141968,
|
| 161 |
+
"rewards/margins": 0.6953709125518799,
|
| 162 |
+
"rewards/rejected": -1.4131637811660767,
|
| 163 |
+
"step": 100
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.49,
|
| 167 |
+
"eval_logits/chosen": -1.2092429399490356,
|
| 168 |
+
"eval_logits/rejected": -0.5798258185386658,
|
| 169 |
+
"eval_logps/chosen": -365.3294677734375,
|
| 170 |
+
"eval_logps/rejected": -413.1092834472656,
|
| 171 |
+
"eval_loss": 0.544935405254364,
|
| 172 |
+
"eval_rewards/accuracies": 0.7160000205039978,
|
| 173 |
+
"eval_rewards/chosen": -0.7798795700073242,
|
| 174 |
+
"eval_rewards/margins": 0.681567370891571,
|
| 175 |
+
"eval_rewards/rejected": -1.4614468812942505,
|
| 176 |
+
"eval_runtime": 384.0193,
|
| 177 |
+
"eval_samples_per_second": 5.208,
|
| 178 |
+
"eval_steps_per_second": 0.651,
|
| 179 |
+
"step": 100
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.54,
|
| 183 |
+
"learning_rate": 2.6072618954988867e-06,
|
| 184 |
+
"logits/chosen": -1.1533472537994385,
|
| 185 |
+
"logits/rejected": -0.3894789218902588,
|
| 186 |
+
"logps/chosen": -369.21466064453125,
|
| 187 |
+
"logps/rejected": -427.78204345703125,
|
| 188 |
+
"loss": 0.5476,
|
| 189 |
+
"rewards/accuracies": 0.6812499761581421,
|
| 190 |
+
"rewards/chosen": -0.81017005443573,
|
| 191 |
+
"rewards/margins": 0.6706447601318359,
|
| 192 |
+
"rewards/rejected": -1.4808146953582764,
|
| 193 |
+
"step": 110
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.59,
|
| 197 |
+
"learning_rate": 2.1790041121336223e-06,
|
| 198 |
+
"logits/chosen": -1.2282952070236206,
|
| 199 |
+
"logits/rejected": -0.6969107985496521,
|
| 200 |
+
"logps/chosen": -362.3310852050781,
|
| 201 |
+
"logps/rejected": -415.934814453125,
|
| 202 |
+
"loss": 0.5101,
|
| 203 |
+
"rewards/accuracies": 0.7281249761581421,
|
| 204 |
+
"rewards/chosen": -0.7746846079826355,
|
| 205 |
+
"rewards/margins": 0.6679913401603699,
|
| 206 |
+
"rewards/rejected": -1.4426759481430054,
|
| 207 |
+
"step": 120
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.64,
|
| 211 |
+
"learning_rate": 1.760183246631777e-06,
|
| 212 |
+
"logits/chosen": -1.172527551651001,
|
| 213 |
+
"logits/rejected": -0.597676694393158,
|
| 214 |
+
"logps/chosen": -354.2320556640625,
|
| 215 |
+
"logps/rejected": -422.12841796875,
|
| 216 |
+
"loss": 0.5083,
|
| 217 |
+
"rewards/accuracies": 0.7406250238418579,
|
| 218 |
+
"rewards/chosen": -0.6290267109870911,
|
| 219 |
+
"rewards/margins": 0.8170250654220581,
|
| 220 |
+
"rewards/rejected": -1.446051836013794,
|
| 221 |
+
"step": 130
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.69,
|
| 225 |
+
"learning_rate": 1.3631121611097364e-06,
|
| 226 |
+
"logits/chosen": -1.188578486442566,
|
| 227 |
+
"logits/rejected": -0.816562294960022,
|
| 228 |
+
"logps/chosen": -359.2029113769531,
|
| 229 |
+
"logps/rejected": -407.4598693847656,
|
| 230 |
+
"loss": 0.5267,
|
| 231 |
+
"rewards/accuracies": 0.715624988079071,
|
| 232 |
+
"rewards/chosen": -0.699051558971405,
|
| 233 |
+
"rewards/margins": 0.6616160273551941,
|
| 234 |
+
"rewards/rejected": -1.3606675863265991,
|
| 235 |
+
"step": 140
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.73,
|
| 239 |
+
"learning_rate": 9.994642986290797e-07,
|
| 240 |
+
"logits/chosen": -1.323190450668335,
|
| 241 |
+
"logits/rejected": -0.7437389492988586,
|
| 242 |
+
"logps/chosen": -343.61151123046875,
|
| 243 |
+
"logps/rejected": -402.93597412109375,
|
| 244 |
+
"loss": 0.5295,
|
| 245 |
+
"rewards/accuracies": 0.746874988079071,
|
| 246 |
+
"rewards/chosen": -0.6734719276428223,
|
| 247 |
+
"rewards/margins": 0.7382540702819824,
|
| 248 |
+
"rewards/rejected": -1.4117259979248047,
|
| 249 |
+
"step": 150
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.78,
|
| 253 |
+
"learning_rate": 6.799304971075383e-07,
|
| 254 |
+
"logits/chosen": -1.250248670578003,
|
| 255 |
+
"logits/rejected": -0.6498929262161255,
|
| 256 |
+
"logps/chosen": -375.78753662109375,
|
| 257 |
+
"logps/rejected": -424.0221252441406,
|
| 258 |
+
"loss": 0.5198,
|
| 259 |
+
"rewards/accuracies": 0.7281249761581421,
|
| 260 |
+
"rewards/chosen": -0.7862330675125122,
|
| 261 |
+
"rewards/margins": 0.6770610213279724,
|
| 262 |
+
"rewards/rejected": -1.463294267654419,
|
| 263 |
+
"step": 160
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 0.83,
|
| 267 |
+
"learning_rate": 4.1390469071538183e-07,
|
| 268 |
+
"logits/chosen": -1.101661205291748,
|
| 269 |
+
"logits/rejected": -0.5704807043075562,
|
| 270 |
+
"logps/chosen": -360.23687744140625,
|
| 271 |
+
"logps/rejected": -435.20220947265625,
|
| 272 |
+
"loss": 0.5199,
|
| 273 |
+
"rewards/accuracies": 0.765625,
|
| 274 |
+
"rewards/chosen": -0.7304506897926331,
|
| 275 |
+
"rewards/margins": 0.9151493310928345,
|
| 276 |
+
"rewards/rejected": -1.6455999612808228,
|
| 277 |
+
"step": 170
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 0.88,
|
| 281 |
+
"learning_rate": 2.092077387824884e-07,
|
| 282 |
+
"logits/chosen": -1.2315480709075928,
|
| 283 |
+
"logits/rejected": -0.714310348033905,
|
| 284 |
+
"logps/chosen": -371.4505310058594,
|
| 285 |
+
"logps/rejected": -436.5375061035156,
|
| 286 |
+
"loss": 0.5269,
|
| 287 |
+
"rewards/accuracies": 0.71875,
|
| 288 |
+
"rewards/chosen": -0.7898632287979126,
|
| 289 |
+
"rewards/margins": 0.7547814249992371,
|
| 290 |
+
"rewards/rejected": -1.5446445941925049,
|
| 291 |
+
"step": 180
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 0.93,
|
| 295 |
+
"learning_rate": 7.185750133542168e-08,
|
| 296 |
+
"logits/chosen": -1.1423877477645874,
|
| 297 |
+
"logits/rejected": -0.7209599018096924,
|
| 298 |
+
"logps/chosen": -351.42633056640625,
|
| 299 |
+
"logps/rejected": -417.2574768066406,
|
| 300 |
+
"loss": 0.5275,
|
| 301 |
+
"rewards/accuracies": 0.703125,
|
| 302 |
+
"rewards/chosen": -0.8558732867240906,
|
| 303 |
+
"rewards/margins": 0.733085572719574,
|
| 304 |
+
"rewards/rejected": -1.588958978652954,
|
| 305 |
+
"step": 190
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.98,
|
| 309 |
+
"learning_rate": 5.891920784984184e-09,
|
| 310 |
+
"logits/chosen": -1.2168794870376587,
|
| 311 |
+
"logits/rejected": -0.6400166749954224,
|
| 312 |
+
"logps/chosen": -360.48065185546875,
|
| 313 |
+
"logps/rejected": -428.3067932128906,
|
| 314 |
+
"loss": 0.4925,
|
| 315 |
+
"rewards/accuracies": 0.778124988079071,
|
| 316 |
+
"rewards/chosen": -0.7727445960044861,
|
| 317 |
+
"rewards/margins": 0.8154066205024719,
|
| 318 |
+
"rewards/rejected": -1.5881513357162476,
|
| 319 |
+
"step": 200
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 0.98,
|
| 323 |
+
"eval_logits/chosen": -1.180118203163147,
|
| 324 |
+
"eval_logits/rejected": -0.6068284511566162,
|
| 325 |
+
"eval_logps/chosen": -367.6444396972656,
|
| 326 |
+
"eval_logps/rejected": -415.7151794433594,
|
| 327 |
+
"eval_loss": 0.5366359949111938,
|
| 328 |
+
"eval_rewards/accuracies": 0.7319999933242798,
|
| 329 |
+
"eval_rewards/chosen": -0.8030293583869934,
|
| 330 |
+
"eval_rewards/margins": 0.6844774484634399,
|
| 331 |
+
"eval_rewards/rejected": -1.4875068664550781,
|
| 332 |
+
"eval_runtime": 384.016,
|
| 333 |
+
"eval_samples_per_second": 5.208,
|
| 334 |
+
"eval_steps_per_second": 0.651,
|
| 335 |
+
"step": 200
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 1.0,
|
| 339 |
+
"step": 204,
|
| 340 |
+
"total_flos": 0.0,
|
| 341 |
+
"train_loss": 0.54004861034599,
|
| 342 |
+
"train_runtime": 9618.5725,
|
| 343 |
+
"train_samples_per_second": 2.717,
|
| 344 |
+
"train_steps_per_second": 0.021
|
| 345 |
+
}
|
| 346 |
+
],
|
| 347 |
+
"logging_steps": 10,
|
| 348 |
+
"max_steps": 204,
|
| 349 |
+
"num_input_tokens_seen": 0,
|
| 350 |
+
"num_train_epochs": 1,
|
| 351 |
+
"save_steps": 100,
|
| 352 |
+
"total_flos": 0.0,
|
| 353 |
+
"train_batch_size": 2,
|
| 354 |
+
"trial_name": null,
|
| 355 |
+
"trial_params": null
|
| 356 |
+
}
|