metadata
license: other
base_model: deepseek-ai/deepseek-coder-1.3b-base
tags:
- trl
- kto
- generated_from_trainer
model-index:
- name: nkto1
results: []
nkto1
This model is a fine-tuned version of deepseek-ai/deepseek-coder-1.3b-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1678
- Eval/rewards/chosen: 2.6953
- Eval/logps/chosen: -115.6411
- Eval/rewards/rejected: -9.9410
- Eval/logps/rejected: -280.4737
- Eval/rewards/margins: 12.6363
- Eval/kl: 9.2011
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-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- lr_scheduler_warmup_steps: 200
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | |
|---|---|---|---|---|
| 0.2054 | 0.9143 | 100 | 0.2561 | 17.1606 |
| 0.0438 | 1.8286 | 200 | 0.2053 | 14.1969 |
| 0.0253 | 2.7429 | 300 | 0.1678 | 9.2011 |
Framework versions
- Transformers 4.43.0.dev0
- Pytorch 2.2.2+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1