metadata
base_model: meta-llama/Meta-Llama-3-8B-Instruct
library_name: peft
license: llama3
tags:
- trl
- kto
- generated_from_trainer
model-index:
- name: llama3_false_positives_0609_KTO_hp_screening_seeds
results: []
llama3_false_positives_0609_KTO_hp_screening_seeds
This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B-Instruct on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.7271
- Eval/rewards/chosen: 0.0966
- Eval/logps/chosen: -206.6140
- Eval/rewards/rejected: 0.0752
- Eval/logps/rejected: -220.8250
- Eval/rewards/margins: 0.0214
- Eval/kl: 0.8580
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: 2
- seed: 9012
- gradient_accumulation_steps: 8
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 4.0
Training results
| Training Loss | Epoch | Step | Validation Loss | |
|---|---|---|---|---|
| 0.5994 | 0.9412 | 12 | 0.7301 | 0.1561 |
| 0.6 | 1.9608 | 25 | 0.7256 | 0.6149 |
| 0.5831 | 2.9804 | 38 | 0.7234 | 0.8062 |
| 0.6595 | 3.7647 | 48 | 0.7271 | 0.8580 |
Framework versions
- PEFT 0.11.1
- Transformers 4.44.0
- Pytorch 2.2.0
- Datasets 2.20.0
- Tokenizers 0.19.1