Hugging Face's logo

zhangjie6
/
Llama-3-8B-Instruct-qlora-adapter

PEFT
Safetensors
trl
sft
Generated from Trainer
Model card Files Community
Llama-3-8B-Instruct-qlora-adapter / README.md
zhangjie6's picture
zhangjie6
zhangjie6/Llama-3-8B-Instruct-qlora-adapter
d7223d0 verified
preview code
|
raw
history blame contribute delete
1.9 kB
metadata
base_model: meta-llama/Meta-Llama-3-8B-Instruct
library_name: peft
tags:
  - trl
  - sft
  - generated_from_trainer
model-index:
  - name: Llama-3-8B-Instruct-qlora-adapter
    results: []

Visualize in Weights & Biases

Llama-3-8B-Instruct-qlora-adapter

This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B-Instruct on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1380

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: 0.0001
  • train_batch_size: 6
  • eval_batch_size: 6
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 24
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 1
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
0.8846 0.2395 10 0.5030
0.3505 0.4790 20 0.1996
0.1636 0.7186 30 0.1449
0.1429 0.9581 40 0.1380

Framework versions

  • PEFT 0.11.1
  • Transformers 4.42.3
  • Pytorch 2.1.2
  • Datasets 2.20.0
  • Tokenizers 0.19.1