BroBro87's picture
End of training
988e0c2 verified
|
raw
history blame
2.28 kB
metadata
license: other
library_name: peft
tags:
  - trl
  - sft
  - generated_from_trainer
base_model: stabilityai/stablelm-zephyr-3b
model-index:
  - name: CloudflareTunedModel
    results: []

CloudflareTunedModel

This model is a fine-tuned version of stabilityai/stablelm-zephyr-3b on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 3.4473

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: 4
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 10
  • total_train_batch_size: 40
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_ratio: 0.05
  • num_epochs: 30

Training results

Training Loss Epoch Step Validation Loss
No log 0.56 1 2.9512
No log 1.67 3 2.9727
No log 2.78 5 3.2578
No log 3.89 7 2.9238
No log 5.0 9 3.3867
0.916 5.56 10 3.3066
0.916 6.67 12 3.2090
0.916 7.78 14 3.4980
0.916 8.89 16 3.5098
0.916 10.0 18 3.4434
0.916 10.56 19 3.4375
0.1354 11.67 21 3.4238
0.1354 12.78 23 3.4336
0.1354 13.89 25 3.4473
0.1354 15.0 27 3.4492
0.1354 15.56 28 3.4492
0.0754 16.67 30 3.4473

Framework versions

  • PEFT 0.9.0
  • Transformers 4.38.2
  • Pytorch 2.1.0+cu121
  • Datasets 2.18.0
  • Tokenizers 0.15.2