Hugging Face's logo

liswei
/
Taiwan-ELM-1_1B-Instruct

Text Generation
Transformers
TensorBoard
Safetensors
Chinese
openelm
conversational
custom_code
Model card Files Metrics Community
Taiwan-ELM-1_1B-Instruct / README.md
liswei's picture
liswei
End of training
727b22b verified
preview code
|
raw
history blame
No virus
1.58 kB
metadata
base_model: liswei/OpenELM-1_1B-zh-cp
tags:
  - llama-factory
  - full
  - generated_from_trainer
model-index:
  - name: OpenELM-1_1B-zh-sft
    results: []

OpenELM-1_1B-zh-sft

This model is a fine-tuned version of liswei/OpenELM-1_1B-zh-cp on the en2tw-alignment-sft and the TaiwanChat datasets. It achieves the following results on the evaluation set:

  • Loss: 1.3355

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: 0.0001
  • train_batch_size: 12
  • eval_batch_size: 12
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 8
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 384
  • total_eval_batch_size: 96
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 1.0

Training results

Training Loss Epoch Step Validation Loss
1.4543 0.3527 500 1.4672
1.3683 0.7053 1000 1.3570

Framework versions

  • Transformers 4.41.1
  • Pytorch 2.3.0+cu121
  • Datasets 2.19.1
  • Tokenizers 0.19.1