EdBerg/Baha_2A

1---2library_name: peft3license: llama34base_model: meta-llama/Meta-Llama-3-8B-Instruct5tags:6- base_model:adapter:meta-llama/Meta-Llama-3-8B-Instruct7- lora8- sft9- transformers10- trl11pipeline_tag: text-generation12model-index:13- name: Baha_2A14  results: []15---16 17<!-- This model card has been generated automatically according to the information the Trainer had access to. You18should probably proofread and complete it, then remove this comment. -->19 20# Baha_2A21 22This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on an unknown dataset.23 24## Model description25 26More information needed27 28## Intended uses & limitations29 30More information needed31 32## Training and evaluation data33 34More information needed35 36## Training procedure37 38### Training hyperparameters39 40The following hyperparameters were used during training:41- learning_rate: 0.000242- train_batch_size: 443- eval_batch_size: 844- seed: 4245- gradient_accumulation_steps: 446- total_train_batch_size: 1647- optimizer: Use OptimizerNames.PAGED_ADAMW with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments48- lr_scheduler_type: cosine49- lr_scheduler_warmup_ratio: 0.0350- training_steps: 50051- mixed_precision_training: Native AMP52 53### Training results54 55 56 57### Framework versions58 59- PEFT 0.18.1.dev060- Transformers 4.57.261- Pytorch 2.9.0+cu12662- Datasets 4.0.063- Tokenizers 0.22.1