webagent / webrl-llama /README.md
Hizy's picture
Upload folder using huggingface_hub
c60f0f4 verified
metadata
library_name: transformers
license: other
base_model: Meta-Llama/Meta-Llama-3.1-8B-Instruct
tags:
  - llama-factory
  - full
  - generated_from_trainer
model-index:
  - name: web_policy_sft
    results: []

web_policy_sft

This model is a fine-tuned version of /ceph/home/muhan01/huggingfacemodels/Meta-Llama-3.1-8B-Instruct on the web_policy_sft dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1971

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
  • distributed_type: multi-GPU
  • num_devices: 4
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 16
  • total_eval_batch_size: 4
  • optimizer: Use OptimizerNames.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: 3.0

Training results

Training Loss Epoch Step Validation Loss
0.2843 0.9394 500 0.2385
0.1062 1.8774 1000 0.1962
0.0573 2.8154 1500 0.1971

Framework versions

  • Transformers 4.57.1
  • Pytorch 2.7.1+cu126
  • Datasets 4.0.0
  • Tokenizers 0.22.1