Hugging Face's logo

Angelectronic
/
gemma-QA-ViMMRC-Squad-v1.1

PEFT
TensorBoard
Safetensors
trl
sft
unsloth
Generated from Trainer
Model card Files Metrics Community
gemma-QA-ViMMRC-Squad-v1.1 / README.md
Angelectronic's picture
Angelectronic
End of training
7492cba verified
preview code
|
raw
history blame
2.12 kB
metadata
license: apache-2.0
library_name: peft
tags:
  - trl
  - sft
  - unsloth
  - generated_from_trainer
base_model: unsloth/gemma-1.1-7b-it-bnb-4bit
model-index:
  - name: gemma-QA-ViMMRC-Squad-v1.1
    results: []

gemma-QA-ViMMRC-Squad-v1.1

This model is a fine-tuned version of unsloth/gemma-1.1-7b-it-bnb-4bit on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 3.3372

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: 16
  • eval_batch_size: 8
  • seed: 3407
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 64
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_steps: 5
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss
0.8 0.2307 320 1.9584
0.4031 0.4614 640 2.0371
0.4 0.6921 960 2.1358
0.4 0.9229 1280 2.2552
0.2328 1.1536 1600 2.4241
0.2 1.3843 1920 2.5637
0.2 1.6150 2240 2.7250
0.1117 1.8457 2560 2.8899
0.1008 2.0764 2880 3.1551
0.0578 2.3071 3200 3.2185
0.0566 2.5379 3520 3.3025
0.0555 2.7686 3840 3.3309
0.0516 2.9993 4160 3.3372

Framework versions

  • PEFT 0.10.0
  • Transformers 4.40.2
  • Pytorch 2.3.0
  • Datasets 2.19.1
  • Tokenizers 0.19.1