code-python / README.md
amztheory's picture
Model save
6653673 verified
metadata
license: apache-2.0
library_name: peft
tags:
  - trl
  - sft
  - generated_from_trainer
base_model: tiiuae/falcon-7b
model-index:
  - name: code-python
    results: []

code-python

This model is a fine-tuned version of tiiuae/falcon-7b on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 1.1593

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: 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: cosine
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss
1.4825 0.3183 30 1.5136
1.4361 0.6366 60 1.4485
1.3649 0.9549 90 1.3628
1.3254 1.2732 120 1.2862
1.1969 1.5915 150 1.2227
1.2245 1.9098 180 1.1776
1.1604 2.2281 210 1.1626
1.1737 2.5464 240 1.1594
1.1012 2.8647 270 1.1593

Framework versions

  • PEFT 0.11.1
  • Transformers 4.41.0
  • Pytorch 2.3.0+cu121
  • Datasets 2.19.1
  • Tokenizers 0.19.1