|
--- |
|
license: llama2 |
|
base_model: TheBloke/Xwin-LM-70B-V0.1-GPTQ |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: Xwin70b_fans |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# Xwin70b_fans |
|
|
|
This model is a fine-tuned version of [TheBloke/Xwin-LM-70B-V0.1-GPTQ](https://huggingface.co/TheBloke/Xwin-LM-70B-V0.1-GPTQ) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.0073 |
|
|
|
## 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: 0.0004 |
|
- train_batch_size: 1 |
|
- eval_batch_size: 1 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 2 |
|
- training_steps: 120 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 1.8351 | 0.02 | 10 | 1.4544 | |
|
| 1.2903 | 0.04 | 20 | 1.2835 | |
|
| 1.1891 | 0.07 | 30 | 1.1606 | |
|
| 1.0652 | 0.09 | 40 | 1.1180 | |
|
| 1.0889 | 0.11 | 50 | 1.0877 | |
|
| 1.0931 | 0.13 | 60 | 1.0617 | |
|
| 0.981 | 0.15 | 70 | 1.0435 | |
|
| 0.9941 | 0.17 | 80 | 1.0276 | |
|
| 0.9802 | 0.2 | 90 | 1.0218 | |
|
| 1.1242 | 0.22 | 100 | 1.0135 | |
|
| 1.0687 | 0.24 | 110 | 1.0111 | |
|
| 0.9263 | 0.26 | 120 | 1.0073 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.35.2 |
|
- Pytorch 2.1.0+cu118 |
|
- Datasets 2.15.0 |
|
- Tokenizers 0.15.0 |
|
|
