metadata
license: llama3
base_model: tsavage68/Summary_L3_1000steps_1e7rate_SFT2
tags:
- trl
- dpo
- generated_from_trainer
model-index:
- name: Summary_L3_150steps_1e8rate_01beta_CSFTDPO
results: []
Summary_L3_150steps_1e8rate_01beta_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: 0.6919
- Rewards/chosen: 0.0011
- Rewards/rejected: -0.0016
- Rewards/accuracies: 0.0800
- Rewards/margins: 0.0027
- Logps/rejected: -15.2799
- Logps/chosen: -9.3721
- Logits/rejected: -1.0959
- Logits/chosen: -1.0973
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-08
- 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: 150
Training results
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.6914 | 0.2004 | 50 | 0.6919 | 0.0004 | -0.0022 | 0.0900 | 0.0026 | -15.2856 | -9.3787 | -1.0954 | -1.0968 |
| 0.6938 | 0.4008 | 100 | 0.6918 | 0.0000 | -0.0027 | 0.1050 | 0.0027 | -15.2908 | -9.3826 | -1.0961 | -1.0975 |
| 0.6936 | 0.6012 | 150 | 0.6919 | 0.0011 | -0.0016 | 0.0800 | 0.0027 | -15.2799 | -9.3721 | -1.0959 | -1.0973 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.0.0+cu117
- Datasets 2.20.0
- Tokenizers 0.19.1