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: 0.6174
Eval/rewards/chosen: 3.9755
Eval/logps/chosen: -167.8246
Eval/rewards/rejected: 3.6477
Eval/logps/rejected: -185.0994
Eval/rewards/margins: 0.3278
Eval/kl: 30.1289

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: 0.0001
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: 6.0

Training results

Training Loss	Epoch	Step	Validation Loss
0.5421	0.9412	12	0.6354	16.3097
0.3533	1.9608	25	0.6275	23.1842
0.1385	2.9804	38	0.6259	27.3092
0.1556	4.0	51	0.6243	30.4607
0.1375	4.9412	63	0.6181	30.1272
0.303	5.6471	72	0.6174	30.1289

Framework versions

PEFT 0.11.1
Transformers 4.44.0
Pytorch 2.2.0
Datasets 2.20.0
Tokenizers 0.19.1