simpo-lora-2
This model is a fine-tuned version of hatakeyama-llm-team/with_halcination_little_codes_ck5200 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5415
- Rewards/chosen: -5.7237
- Rewards/rejected: -12.8559
- Rewards/accuracies: 0.7368
- Rewards/margins: 7.1321
- Logps/rejected: -5.1424
- Logps/chosen: -2.2895
- Logits/rejected: -0.5526
- Logits/chosen: -0.5626
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: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 64
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- 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 |
---|---|---|---|---|---|---|---|---|---|---|---|
0.7791 | 0.1600 | 100 | 0.9134 | -4.7310 | -6.5490 | 0.7037 | 1.8179 | -2.6196 | -1.8924 | -0.1767 | 0.0399 |
0.4822 | 0.3200 | 200 | 0.5821 | -5.4149 | -11.9630 | 0.7349 | 6.5482 | -4.7852 | -2.1659 | 0.0142 | 0.0556 |
0.5591 | 0.4800 | 300 | 0.5573 | -5.5227 | -12.0727 | 0.7417 | 6.5500 | -4.8291 | -2.2091 | -0.6007 | -0.5349 |
0.6293 | 0.6400 | 400 | 0.5447 | -5.8900 | -14.0923 | 0.7378 | 8.2023 | -5.6369 | -2.3560 | -0.1891 | -0.2711 |
0.5645 | 0.8000 | 500 | 0.5426 | -5.7599 | -13.0285 | 0.7368 | 7.2686 | -5.2114 | -2.3040 | -0.5393 | -0.5567 |
0.4763 | 0.9600 | 600 | 0.5415 | -5.7237 | -12.8559 | 0.7368 | 7.1321 | -5.1424 | -2.2895 | -0.5526 | -0.5626 |
Framework versions
- PEFT 0.11.1
- Transformers 4.41.1
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
- Downloads last month
- 2
Model tree for misdelivery/simpo-lora-2
Adapter
this model