Hugging Face's logo

allen0909
/
ROE_QA_Breeze-7B-Instruct-v1_0_Q100_80_20

PEFT
Safetensors
mistral
trl
sft
Generated from Trainer
Model card Files Community
ROE_QA_Breeze-7B-Instruct-v1_0_Q100_80_20 / README.md
allen0909's picture
allen0909
End of training
1292456 verified
preview code
|
raw
history blame contribute delete
1.33 kB
metadata
license: apache-2.0
library_name: peft
tags:
  - trl
  - sft
  - generated_from_trainer
base_model: MediaTek-Research/Breeze-7B-Instruct-v1_0
model-index:
  - name: ROE_QA_Breeze-7B-Instruct-v1_0_1
    results: []

ROE_QA_Breeze-7B-Instruct-v1_0_1

This model is a fine-tuned version of MediaTek-Research/Breeze-7B-Instruct-v1_0 on the None 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: 0.0003903677440817261
  • train_batch_size: 4
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 16
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: constant
  • num_epochs: 5
  • mixed_precision_training: Native AMP

Training results

Framework versions

  • PEFT 0.12.1.dev0
  • Transformers 4.44.2
  • Pytorch 2.3.0+cu121
  • Datasets 2.19.1
  • Tokenizers 0.19.1