metadata
library_name: transformers
license: gemma
base_model: google/gemma-7b
tags:
- alignment-handbook
- trl
- orpo
- generated_from_trainer
datasets:
- argilla/dpo-mix-7k
model-index:
- name: gemma-7b-orpo
results: []
gemma-7b-orpo
This model is a fine-tuned version of google/gemma-7b on the argilla/dpo-mix-7k dataset. It achieves the following results on the evaluation set:
- Loss: 1.4556
- Rewards/chosen: -0.0513
- Rewards/rejected: -0.0589
- Rewards/accuracies: 0.5108
- Rewards/margins: 0.0076
- Logps/rejected: -1.1787
- Logps/chosen: -1.0268
- Logits/rejected: 312.9670
- Logits/chosen: 340.5321
- Nll Loss: 1.4096
- Log Odds Ratio: -0.6928
- Log Odds Chosen: 0.2398
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: 5e-06
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- total_train_batch_size: 4
- total_eval_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen | Nll Loss | Log Odds Ratio | Log Odds Chosen |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1.3423 | 1.0 | 1259 | 1.4556 | -0.0513 | -0.0589 | 0.5108 | 0.0076 | -1.1787 | -1.0268 | 312.9670 | 340.5321 | 1.4096 | -0.6928 | 0.2398 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.0+cu121
- Datasets 3.0.0
- Tokenizers 0.19.1