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_real_rank4_all_iter_3
|
| 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_real_rank4_all_iter_3
|
| 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: 0.5483
|
| 22 |
+
- Rewards/chosen: -0.5406
|
| 23 |
+
- Rewards/rejected: -1.1226
|
| 24 |
+
- Rewards/accuracies: 0.6960
|
| 25 |
+
- Rewards/margins: 0.5820
|
| 26 |
+
- Logps/rejected: -399.9344
|
| 27 |
+
- Logps/chosen: -360.6637
|
| 28 |
+
- Logits/rejected: 0.5949
|
| 29 |
+
- Logits/chosen: -0.0563
|
| 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 |
+
| 0.5359 | 0.49 | 100 | 0.5606 | -0.4171 | -0.9562 | 0.7000 | 0.5391 | -383.2901 | -348.3135 | 0.5653 | -0.0477 |
|
| 67 |
+
| 0.5121 | 0.98 | 200 | 0.5483 | -0.5406 | -1.1226 | 0.6960 | 0.5820 | -399.9344 | -360.6637 | 0.5949 | -0.0563 |
|
| 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": 0.5619814822486803,
|
| 4 |
+
"train_runtime": 9621.729,
|
| 5 |
+
"train_samples": 26135,
|
| 6 |
+
"train_samples_per_second": 2.716,
|
| 7 |
+
"train_steps_per_second": 0.021
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.5619814822486803,
|
| 4 |
+
"train_runtime": 9621.729,
|
| 5 |
+
"train_samples": 26135,
|
| 6 |
+
"train_samples_per_second": 2.716,
|
| 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": -2.530543327331543,
|
| 15 |
+
"logits/rejected": -2.3806393146514893,
|
| 16 |
+
"logps/chosen": -277.43206787109375,
|
| 17 |
+
"logps/rejected": -259.5929870605469,
|
| 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": -2.3367958068847656,
|
| 29 |
+
"logits/rejected": -2.3201417922973633,
|
| 30 |
+
"logps/chosen": -237.50930786132812,
|
| 31 |
+
"logps/rejected": -259.6923828125,
|
| 32 |
+
"loss": 0.6911,
|
| 33 |
+
"rewards/accuracies": 0.5138888955116272,
|
| 34 |
+
"rewards/chosen": -0.0002860191452782601,
|
| 35 |
+
"rewards/margins": 0.004583699628710747,
|
| 36 |
+
"rewards/rejected": -0.004869719035923481,
|
| 37 |
+
"step": 10
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.1,
|
| 41 |
+
"learning_rate": 4.761904761904762e-06,
|
| 42 |
+
"logits/chosen": -2.2300033569335938,
|
| 43 |
+
"logits/rejected": -2.062291383743286,
|
| 44 |
+
"logps/chosen": -322.0083312988281,
|
| 45 |
+
"logps/rejected": -298.00164794921875,
|
| 46 |
+
"loss": 0.6623,
|
| 47 |
+
"rewards/accuracies": 0.6937500238418579,
|
| 48 |
+
"rewards/chosen": -0.05355127900838852,
|
| 49 |
+
"rewards/margins": 0.07823377847671509,
|
| 50 |
+
"rewards/rejected": -0.1317850649356842,
|
| 51 |
+
"step": 20
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.15,
|
| 55 |
+
"learning_rate": 4.970219740227693e-06,
|
| 56 |
+
"logits/chosen": -1.7139427661895752,
|
| 57 |
+
"logits/rejected": -1.5317604541778564,
|
| 58 |
+
"logps/chosen": -302.0789489746094,
|
| 59 |
+
"logps/rejected": -304.89666748046875,
|
| 60 |
+
"loss": 0.6183,
|
| 61 |
+
"rewards/accuracies": 0.65625,
|
| 62 |
+
"rewards/chosen": -0.13164304196834564,
|
| 63 |
+
"rewards/margins": 0.1831807643175125,
|
| 64 |
+
"rewards/rejected": -0.31482380628585815,
|
| 65 |
+
"step": 30
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.2,
|
| 69 |
+
"learning_rate": 4.868186180746792e-06,
|
| 70 |
+
"logits/chosen": -1.342703104019165,
|
| 71 |
+
"logits/rejected": -1.0153623819351196,
|
| 72 |
+
"logps/chosen": -317.6343688964844,
|
| 73 |
+
"logps/rejected": -340.515625,
|
| 74 |
+
"loss": 0.6019,
|
| 75 |
+
"rewards/accuracies": 0.6937500238418579,
|
| 76 |
+
"rewards/chosen": -0.03232262283563614,
|
| 77 |
+
"rewards/margins": 0.30836719274520874,
|
| 78 |
+
"rewards/rejected": -0.34068983793258667,
|
| 79 |
+
"step": 40
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.24,
|
| 83 |
+
"learning_rate": 4.696530612642871e-06,
|
| 84 |
+
"logits/chosen": -0.4007461667060852,
|
| 85 |
+
"logits/rejected": -0.062369205057621,
|
| 86 |
+
"logps/chosen": -344.5245056152344,
|
| 87 |
+
"logps/rejected": -381.02655029296875,
|
| 88 |
+
"loss": 0.5788,
|
| 89 |
+
"rewards/accuracies": 0.7124999761581421,
|
| 90 |
+
"rewards/chosen": -0.4034351408481598,
|
| 91 |
+
"rewards/margins": 0.35783329606056213,
|
| 92 |
+
"rewards/rejected": -0.7612683773040771,
|
| 93 |
+
"step": 50
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.29,
|
| 97 |
+
"learning_rate": 4.460299516441777e-06,
|
| 98 |
+
"logits/chosen": -0.4702862799167633,
|
| 99 |
+
"logits/rejected": -0.10355714708566666,
|
| 100 |
+
"logps/chosen": -309.2970886230469,
|
| 101 |
+
"logps/rejected": -356.2215881347656,
|
| 102 |
+
"loss": 0.5714,
|
| 103 |
+
"rewards/accuracies": 0.715624988079071,
|
| 104 |
+
"rewards/chosen": -0.17937391996383667,
|
| 105 |
+
"rewards/margins": 0.42016029357910156,
|
| 106 |
+
"rewards/rejected": -0.5995342135429382,
|
| 107 |
+
"step": 60
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.34,
|
| 111 |
+
"learning_rate": 4.1664378205239085e-06,
|
| 112 |
+
"logits/chosen": -0.14899344742298126,
|
| 113 |
+
"logits/rejected": 0.4177340567111969,
|
| 114 |
+
"logps/chosen": -340.29766845703125,
|
| 115 |
+
"logps/rejected": -379.6929626464844,
|
| 116 |
+
"loss": 0.5634,
|
| 117 |
+
"rewards/accuracies": 0.684374988079071,
|
| 118 |
+
"rewards/chosen": -0.4960289001464844,
|
| 119 |
+
"rewards/margins": 0.4510241448879242,
|
| 120 |
+
"rewards/rejected": -0.9470531344413757,
|
| 121 |
+
"step": 70
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.39,
|
| 125 |
+
"learning_rate": 3.8235847280454626e-06,
|
| 126 |
+
"logits/chosen": -0.14341413974761963,
|
| 127 |
+
"logits/rejected": 0.4753722548484802,
|
| 128 |
+
"logps/chosen": -334.1756286621094,
|
| 129 |
+
"logps/rejected": -374.01556396484375,
|
| 130 |
+
"loss": 0.527,
|
| 131 |
+
"rewards/accuracies": 0.715624988079071,
|
| 132 |
+
"rewards/chosen": -0.3731788992881775,
|
| 133 |
+
"rewards/margins": 0.5048251152038574,
|
| 134 |
+
"rewards/rejected": -0.8780039548873901,
|
| 135 |
+
"step": 80
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.44,
|
| 139 |
+
"learning_rate": 3.441819734087963e-06,
|
| 140 |
+
"logits/chosen": -0.20099422335624695,
|
| 141 |
+
"logits/rejected": 0.4678555130958557,
|
| 142 |
+
"logps/chosen": -348.76397705078125,
|
| 143 |
+
"logps/rejected": -384.9765319824219,
|
| 144 |
+
"loss": 0.5453,
|
| 145 |
+
"rewards/accuracies": 0.7406250238418579,
|
| 146 |
+
"rewards/chosen": -0.3222987651824951,
|
| 147 |
+
"rewards/margins": 0.5874095559120178,
|
| 148 |
+
"rewards/rejected": -0.9097083210945129,
|
| 149 |
+
"step": 90
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.49,
|
| 153 |
+
"learning_rate": 3.0323662998460396e-06,
|
| 154 |
+
"logits/chosen": -0.17653772234916687,
|
| 155 |
+
"logits/rejected": 0.5543693900108337,
|
| 156 |
+
"logps/chosen": -353.01361083984375,
|
| 157 |
+
"logps/rejected": -381.8619689941406,
|
| 158 |
+
"loss": 0.5359,
|
| 159 |
+
"rewards/accuracies": 0.703125,
|
| 160 |
+
"rewards/chosen": -0.41476544737815857,
|
| 161 |
+
"rewards/margins": 0.5261854529380798,
|
| 162 |
+
"rewards/rejected": -0.940950870513916,
|
| 163 |
+
"step": 100
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.49,
|
| 167 |
+
"eval_logits/chosen": -0.04770730063319206,
|
| 168 |
+
"eval_logits/rejected": 0.5653438568115234,
|
| 169 |
+
"eval_logps/chosen": -348.3134765625,
|
| 170 |
+
"eval_logps/rejected": -383.2900695800781,
|
| 171 |
+
"eval_loss": 0.5605703592300415,
|
| 172 |
+
"eval_rewards/accuracies": 0.699999988079071,
|
| 173 |
+
"eval_rewards/chosen": -0.4171307384967804,
|
| 174 |
+
"eval_rewards/margins": 0.5390676259994507,
|
| 175 |
+
"eval_rewards/rejected": -0.9561984539031982,
|
| 176 |
+
"eval_runtime": 383.9702,
|
| 177 |
+
"eval_samples_per_second": 5.209,
|
| 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": -0.024105533957481384,
|
| 185 |
+
"logits/rejected": 0.6808269619941711,
|
| 186 |
+
"logps/chosen": -357.918701171875,
|
| 187 |
+
"logps/rejected": -408.1763610839844,
|
| 188 |
+
"loss": 0.5632,
|
| 189 |
+
"rewards/accuracies": 0.7250000238418579,
|
| 190 |
+
"rewards/chosen": -0.5132647752761841,
|
| 191 |
+
"rewards/margins": 0.5779321193695068,
|
| 192 |
+
"rewards/rejected": -1.0911967754364014,
|
| 193 |
+
"step": 110
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.59,
|
| 197 |
+
"learning_rate": 2.1790041121336223e-06,
|
| 198 |
+
"logits/chosen": -0.00899410154670477,
|
| 199 |
+
"logits/rejected": 0.5235549211502075,
|
| 200 |
+
"logps/chosen": -369.63299560546875,
|
| 201 |
+
"logps/rejected": -408.86859130859375,
|
| 202 |
+
"loss": 0.5253,
|
| 203 |
+
"rewards/accuracies": 0.7281249761581421,
|
| 204 |
+
"rewards/chosen": -0.640425980091095,
|
| 205 |
+
"rewards/margins": 0.5291002988815308,
|
| 206 |
+
"rewards/rejected": -1.1695263385772705,
|
| 207 |
+
"step": 120
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.64,
|
| 211 |
+
"learning_rate": 1.760183246631777e-06,
|
| 212 |
+
"logits/chosen": 0.018746722489595413,
|
| 213 |
+
"logits/rejected": 0.6442402005195618,
|
| 214 |
+
"logps/chosen": -351.2228088378906,
|
| 215 |
+
"logps/rejected": -409.70623779296875,
|
| 216 |
+
"loss": 0.5308,
|
| 217 |
+
"rewards/accuracies": 0.7281249761581421,
|
| 218 |
+
"rewards/chosen": -0.4084150195121765,
|
| 219 |
+
"rewards/margins": 0.7148237824440002,
|
| 220 |
+
"rewards/rejected": -1.1232386827468872,
|
| 221 |
+
"step": 130
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.69,
|
| 225 |
+
"learning_rate": 1.3631121611097364e-06,
|
| 226 |
+
"logits/chosen": 0.052795905619859695,
|
| 227 |
+
"logits/rejected": 0.5363299250602722,
|
| 228 |
+
"logps/chosen": -350.0050354003906,
|
| 229 |
+
"logps/rejected": -389.28875732421875,
|
| 230 |
+
"loss": 0.5368,
|
| 231 |
+
"rewards/accuracies": 0.699999988079071,
|
| 232 |
+
"rewards/chosen": -0.411121129989624,
|
| 233 |
+
"rewards/margins": 0.5764644742012024,
|
| 234 |
+
"rewards/rejected": -0.9875855445861816,
|
| 235 |
+
"step": 140
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.73,
|
| 239 |
+
"learning_rate": 9.994642986290797e-07,
|
| 240 |
+
"logits/chosen": -0.08179731667041779,
|
| 241 |
+
"logits/rejected": 0.6078404784202576,
|
| 242 |
+
"logps/chosen": -340.8910217285156,
|
| 243 |
+
"logps/rejected": -395.11309814453125,
|
| 244 |
+
"loss": 0.5396,
|
| 245 |
+
"rewards/accuracies": 0.737500011920929,
|
| 246 |
+
"rewards/chosen": -0.4558540880680084,
|
| 247 |
+
"rewards/margins": 0.681885302066803,
|
| 248 |
+
"rewards/rejected": -1.1377394199371338,
|
| 249 |
+
"step": 150
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.78,
|
| 253 |
+
"learning_rate": 6.799304971075383e-07,
|
| 254 |
+
"logits/chosen": -0.10812912881374359,
|
| 255 |
+
"logits/rejected": 0.7408884763717651,
|
| 256 |
+
"logps/chosen": -367.9963684082031,
|
| 257 |
+
"logps/rejected": -409.3976135253906,
|
| 258 |
+
"loss": 0.5303,
|
| 259 |
+
"rewards/accuracies": 0.7406250238418579,
|
| 260 |
+
"rewards/chosen": -0.5040684938430786,
|
| 261 |
+
"rewards/margins": 0.6103622317314148,
|
| 262 |
+
"rewards/rejected": -1.1144306659698486,
|
| 263 |
+
"step": 160
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 0.83,
|
| 267 |
+
"learning_rate": 4.1390469071538183e-07,
|
| 268 |
+
"logits/chosen": 0.11164654791355133,
|
| 269 |
+
"logits/rejected": 0.6015251278877258,
|
| 270 |
+
"logps/chosen": -354.7975158691406,
|
| 271 |
+
"logps/rejected": -422.2373046875,
|
| 272 |
+
"loss": 0.5308,
|
| 273 |
+
"rewards/accuracies": 0.7562500238418579,
|
| 274 |
+
"rewards/chosen": -0.47427788376808167,
|
| 275 |
+
"rewards/margins": 0.8326796293258667,
|
| 276 |
+
"rewards/rejected": -1.306957483291626,
|
| 277 |
+
"step": 170
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 0.88,
|
| 281 |
+
"learning_rate": 2.092077387824884e-07,
|
| 282 |
+
"logits/chosen": -0.10529808700084686,
|
| 283 |
+
"logits/rejected": 0.4767494201660156,
|
| 284 |
+
"logps/chosen": -361.30084228515625,
|
| 285 |
+
"logps/rejected": -419.2337951660156,
|
| 286 |
+
"loss": 0.5424,
|
| 287 |
+
"rewards/accuracies": 0.7093750238418579,
|
| 288 |
+
"rewards/chosen": -0.48817354440689087,
|
| 289 |
+
"rewards/margins": 0.6789232492446899,
|
| 290 |
+
"rewards/rejected": -1.167096734046936,
|
| 291 |
+
"step": 180
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 0.93,
|
| 295 |
+
"learning_rate": 7.185750133542168e-08,
|
| 296 |
+
"logits/chosen": 0.029346108436584473,
|
| 297 |
+
"logits/rejected": 0.39755165576934814,
|
| 298 |
+
"logps/chosen": -337.2845764160156,
|
| 299 |
+
"logps/rejected": -398.9493713378906,
|
| 300 |
+
"loss": 0.54,
|
| 301 |
+
"rewards/accuracies": 0.703125,
|
| 302 |
+
"rewards/chosen": -0.5243724584579468,
|
| 303 |
+
"rewards/margins": 0.6791526675224304,
|
| 304 |
+
"rewards/rejected": -1.203525185585022,
|
| 305 |
+
"step": 190
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.98,
|
| 309 |
+
"learning_rate": 5.891920784984184e-09,
|
| 310 |
+
"logits/chosen": -0.1364443600177765,
|
| 311 |
+
"logits/rejected": 0.5197805166244507,
|
| 312 |
+
"logps/chosen": -351.525390625,
|
| 313 |
+
"logps/rejected": -410.6580505371094,
|
| 314 |
+
"loss": 0.5121,
|
| 315 |
+
"rewards/accuracies": 0.778124988079071,
|
| 316 |
+
"rewards/chosen": -0.4877961575984955,
|
| 317 |
+
"rewards/margins": 0.7241436839103699,
|
| 318 |
+
"rewards/rejected": -1.211940050125122,
|
| 319 |
+
"step": 200
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 0.98,
|
| 323 |
+
"eval_logits/chosen": -0.05626865103840828,
|
| 324 |
+
"eval_logits/rejected": 0.5949223637580872,
|
| 325 |
+
"eval_logps/chosen": -360.6636657714844,
|
| 326 |
+
"eval_logps/rejected": -399.93438720703125,
|
| 327 |
+
"eval_loss": 0.5482509732246399,
|
| 328 |
+
"eval_rewards/accuracies": 0.6959999799728394,
|
| 329 |
+
"eval_rewards/chosen": -0.5406328439712524,
|
| 330 |
+
"eval_rewards/margins": 0.5820090770721436,
|
| 331 |
+
"eval_rewards/rejected": -1.122641921043396,
|
| 332 |
+
"eval_runtime": 384.3787,
|
| 333 |
+
"eval_samples_per_second": 5.203,
|
| 334 |
+
"eval_steps_per_second": 0.65,
|
| 335 |
+
"step": 200
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 1.0,
|
| 339 |
+
"step": 204,
|
| 340 |
+
"total_flos": 0.0,
|
| 341 |
+
"train_loss": 0.5619814822486803,
|
| 342 |
+
"train_runtime": 9621.729,
|
| 343 |
+
"train_samples_per_second": 2.716,
|
| 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 |
+
}
|