Dynosaur's picture
End of training
67c3e8c verified
metadata
library_name: transformers
license: llama3
base_model: Dynosaur/llama3-8b-math-sft
tags:
  - alignment-handbook
  - trl
  - sft
  - generated_from_trainer
  - trl
  - sft
  - generated_from_trainer
datasets:
  - hexuan21/math-sft-mix-8-1
model-index:
  - name: llama3-8b-math-sft-mix-8-1
    results: []

llama3-8b-math-sft-mix-8-1

This model is a fine-tuned version of Dynosaur/llama3-8b-math-sft on the hexuan21/math-sft-mix-8-1 dataset.

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: 2e-05
  • train_batch_size: 2
  • eval_batch_size: 8
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 4
  • gradient_accumulation_steps: 8
  • total_train_batch_size: 64
  • total_eval_batch_size: 32
  • optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 2

Training results

Framework versions

  • Transformers 4.46.3
  • Pytorch 2.5.1+cu124
  • Datasets 3.1.0
  • Tokenizers 0.20.3