minhbui's picture
Upload folder using huggingface_hub
15e47ff
metadata
base_model: meta-llama/Llama-2-7b-hf
tags:
  - generated_from_trainer
model-index:
  - name: ckpts/llama2-7b-viettel_v3.2_2epoch
    results: []

Built with Axolotl

ckpts/llama2-7b-viettel_v3.2_2epoch

This model is a fine-tuned version of meta-llama/Llama-2-7b-hf on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3727

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: 0.0002
  • train_batch_size: 2
  • eval_batch_size: 2
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 3
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 24
  • total_eval_batch_size: 6
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_steps: 20
  • num_epochs: 2

Training results

Training Loss Epoch Step Validation Loss
0.4378 0.12 200 0.4331
0.4266 0.24 400 0.4187
0.4199 0.37 600 0.4086
0.4024 0.49 800 0.4016
0.4003 0.61 1000 0.3966
0.3849 0.73 1200 0.3914
0.3814 0.86 1400 0.3865
0.3825 0.98 1600 0.3831
0.3557 1.1 1800 0.3812
0.3531 1.22 2000 0.3789
0.3444 1.35 2200 0.3771
0.3411 1.47 2400 0.3752
0.35 1.59 2600 0.3738
0.3586 1.71 2800 0.3733
0.349 1.84 3000 0.3728
0.357 1.96 3200 0.3727

Framework versions

  • Transformers 4.34.0.dev0
  • Pytorch 2.0.1+cu117
  • Datasets 2.14.4
  • Tokenizers 0.14.0