qwen1.5-1.8b-sft / README.md
lewtun's picture
lewtun HF staff
Add HuggingFaceH4/qwen-1.5-1.8b-sft-v0.0 checkpoint
6eac95c verified
metadata
license: other
base_model: Qwen/Qwen1.5-1.8B
tags:
  - alignment-handbook
  - generated_from_trainer
datasets:
  - HuggingFaceH4/deita-6k-v0-sft
model-index:
  - name: qwen-1.5-1.8b-sft-v0.0
    results: []

qwen-1.5-1.8b-sft-v0.0

This model is a fine-tuned version of Qwen/Qwen1.5-1.8B on the HuggingFaceH4/deita-6k-v0-sft dataset. It achieves the following results on the evaluation set:

  • Loss: 1.0886

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: 16
  • eval_batch_size: 8
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 8
  • total_train_batch_size: 128
  • total_eval_batch_size: 64
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss
1.0963 1.0 237 1.0788
0.9779 2.0 474 1.0728
0.8734 3.0 711 1.0886

Framework versions

  • Transformers 4.38.2
  • Pytorch 2.1.2+cu121
  • Datasets 2.16.1
  • Tokenizers 0.15.1