ndpo2
This model is a fine-tuned version of deepseek-ai/deepseek-coder-1.3b-base on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 0.0429
- Rewards/chosen: 1.0992
- Rewards/rejected: -12.7484
- Rewards/accuracies: 0.9836
- Rewards/margins: 13.8476
- Logps/rejected: -300.4520
- Logps/chosen: -157.4631
- Logits/rejected: -15.0529
- Logits/chosen: -15.3161
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-05
- train_batch_size: 2
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 64
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 200
- num_epochs: 2
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.0841 | 1.7149 | 100 | 0.0429 | 1.0992 | -12.7484 | 0.9836 | 13.8476 | -300.4520 | -157.4631 | -15.0529 | -15.3161 |
Framework versions
- Transformers 4.43.0.dev0
- Pytorch 2.2.2+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1