llama2-lima / README.md
pkarypis's picture
End of training
a4c2cfd verified
metadata
license: llama2
base_model: meta-llama/Llama-2-7b-hf
tags:
  - alignment-handbook
  - trl
  - sft
  - generated_from_trainer
  - trl
  - sft
  - generated_from_trainer
datasets:
  - GAIR/lima
model-index:
  - name: llama2-lima
    results: []

llama2-lima

This model is a fine-tuned version of meta-llama/Llama-2-7b-hf on the GAIR/lima dataset. It achieves the following results on the evaluation set:

  • Loss: 2.5297

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: 8
  • eval_batch_size: 8
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 16
  • total_train_batch_size: 128
  • total_eval_batch_size: 128
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 10.0

Training results

Training Loss Epoch Step Validation Loss
1.9786 1.0 6 2.6180
1.75 2.0 12 1.9988
1.5489 3.0 18 1.9909
1.3631 4.0 24 1.9766
1.0134 5.0 30 2.1003
0.8425 6.0 36 2.1683
0.7051 7.0 42 2.3737
0.594 8.0 48 2.5296
0.5107 9.0 54 2.5203
0.4594 10.0 60 2.5297

Framework versions

  • Transformers 4.38.2
  • Pytorch 2.1.2
  • Datasets 2.14.6
  • Tokenizers 0.15.2