metadata
license: llama3
base_model: tsavage68/Summary_L3_1000steps_1e7rate_SFT2
tags:
- trl
- dpo
- generated_from_trainer
model-index:
- name: Summary_L3_200steps_1e7rate_03beta_CSFTDPO
results: []
Summary_L3_200steps_1e7rate_03beta_CSFTDPO
This model is a fine-tuned version of tsavage68/Summary_L3_1000steps_1e7rate_SFT2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.6121
- Rewards/chosen: 0.0271
- Rewards/rejected: -0.2888
- Rewards/accuracies: 0.1400
- Rewards/margins: 0.3159
- Logps/rejected: -16.2264
- Logps/chosen: -9.2923
- Logits/rejected: -1.0930
- Logits/chosen: -1.0946
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-07
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- training_steps: 200
Training results
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.6831 | 0.2004 | 50 | 0.6816 | 0.0015 | -0.0238 | 0.1300 | 0.0253 | -15.3431 | -9.3779 | -1.0962 | -1.0977 |
| 0.6795 | 0.4008 | 100 | 0.6463 | 0.0093 | -0.1112 | 0.1400 | 0.1205 | -15.6344 | -9.3518 | -1.0932 | -1.0948 |
| 0.6344 | 0.6012 | 150 | 0.6127 | 0.0273 | -0.2824 | 0.1400 | 0.3097 | -16.2051 | -9.2918 | -1.0930 | -1.0947 |
| 0.6204 | 0.8016 | 200 | 0.6121 | 0.0271 | -0.2888 | 0.1400 | 0.3159 | -16.2264 | -9.2923 | -1.0930 | -1.0946 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.0.0+cu117
- Datasets 2.20.0
- Tokenizers 0.19.1