|
--- |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: myBit-Llama2-jp-127M-3 |
|
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. --> |
|
|
|
# myBit-Llama2-jp-127M-3 |
|
|
|
This model is a fine-tuned version of [](https://huggingface.co./) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 3.8396 |
|
|
|
## 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: 0.00024 |
|
- train_batch_size: 96 |
|
- eval_batch_size: 96 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: polynomial |
|
- lr_scheduler_warmup_steps: 5000 |
|
- num_epochs: 1 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
| 6.9048 | 0.05 | 2000 | 4.7602 | |
|
| 4.4421 | 0.1 | 4000 | 4.2117 | |
|
| 4.0625 | 0.15 | 6000 | 3.9227 | |
|
| 3.807 | 0.2 | 8000 | 3.7181 | |
|
| 3.6547 | 0.25 | 10000 | 3.5929 | |
|
| 3.5296 | 0.29 | 12000 | 3.4812 | |
|
| 3.4492 | 0.34 | 14000 | 3.4236 | |
|
| 3.4065 | 0.39 | 16000 | 3.3923 | |
|
| 3.3816 | 0.44 | 18000 | 3.3778 | |
|
| 3.3815 | 0.49 | 20000 | 3.3907 | |
|
| 3.431 | 0.54 | 22000 | 3.4870 | |
|
| 3.5507 | 0.59 | 24000 | 3.5969 | |
|
| 3.6557 | 0.64 | 26000 | 3.6918 | |
|
| 3.715 | 0.69 | 28000 | 3.7377 | |
|
| 3.7646 | 0.74 | 30000 | 3.7620 | |
|
| 3.8005 | 0.79 | 32000 | 3.8221 | |
|
| 3.8288 | 0.83 | 34000 | 3.8550 | |
|
| 3.8552 | 0.88 | 36000 | 3.8449 | |
|
| 3.8591 | 0.93 | 38000 | 3.8483 | |
|
| 3.8452 | 0.98 | 40000 | 3.8396 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.38.2 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
|
|
