FT-DS / README.md
Karun-7's picture
Karun-7/FT-DS
36ab475 verified
metadata
base_model: TheBloke/Mistral-7B-Instruct-v0.2-GPTQ
library_name: peft
license: apache-2.0
tags:
  - generated_from_trainer
model-index:
  - name: FT-DS
    results: []

FT-DS

This model is a fine-tuned version of TheBloke/Mistral-7B-Instruct-v0.2-GPTQ on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 1.2844

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: 9
  • eval_batch_size: 2
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 36
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 2
  • num_epochs: 10
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
0.9542 1.0 1 1.8234
0.954 2.0 2 1.7674
0.9305 3.0 3 1.6566
0.8837 4.0 4 1.5636
0.843 5.0 5 1.4859
0.8086 6.0 6 1.4216
0.7804 7.0 7 1.3689
0.7578 8.0 8 1.3279
0.7405 9.0 9 1.2993
0.7283 10.0 10 1.2844

Framework versions

  • PEFT 0.12.0
  • Transformers 4.44.2
  • Pytorch 2.3.1+cu121
  • Datasets 2.21.0
  • Tokenizers 0.19.1