metadata
license: other
tags:
- generated_from_trainer
base_model: KnutJaegersberg/Qwen-1_8B-Llamafied
model-index:
- name: qwen-1.8b-vi
results: []
Quan-1.8b
Qwen-1.8B finetuned on bilingual English-Vietnamese Data.
Prompt Template
ChatML, same as VinaLlama
<|im_start|>system
Bạn là một trợ lí AI hữu ích. Hãy trả lời người dùng một cách chính xác.
<|im_end|>
<|im_start|>user
Hello world!<|im_end|>
<|im_start|>assistant
This model is a fine-tuned version of KnutJaegersberg/Qwen-1_8B-Llamafied on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.8096
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: 2
- eval_batch_size: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- total_eval_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.8123 | 1.02 | 2356 | 0.8183 |
| 0.7358 | 2.02 | 4713 | 0.7790 |
| 0.6379 | 3.02 | 7071 | 0.7822 |
| 0.5762 | 3.94 | 9252 | 0.8096 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.0.1+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 45.91 |
| AI2 Reasoning Challenge (25-Shot) | 39.08 |
| HellaSwag (10-Shot) | 62.37 |
| MMLU (5-Shot) | 44.09 |
| TruthfulQA (0-shot) | 43.15 |
| Winogrande (5-shot) | 59.27 |
| GSM8k (5-shot) | 27.52 |