|
--- |
|
license: apache-2.0 |
|
tags: |
|
- KoRWKV |
|
- KoAlpaca |
|
model-index: |
|
- name: KoAlpaca-KoRWKV-1.5B |
|
results: [] |
|
language: |
|
- ko |
|
datasets: |
|
- KoAlpaca-v1.0 |
|
pipeline_tag: text-generation |
|
--- |
|
|
|
> 🚧 Note: this repo is under construction, current uploaded version is finetuned version of KoRWKV which is ~20% trained ckpt (with ~31Billion tokens) 🚧 |
|
|
|
# beomi/KoAlpaca-KoRWKV-1.5B (v1.0) |
|
|
|
This model is a fine-tuned version of [KoRWKV-1.5B](https://huggingface.co/beomi/KoRWKV-1.5B) on a KoAlpaca Dataset v1.0 |
|
|
|
Dataset available at [KoAlpaca Github Repository](https://github.com/Beomi/KoAlpaca) |
|
|
|
## Training procedure |
|
|
|
### Train Device |
|
|
|
- A100 80G x2 |
|
- ~2hrs |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 5e-05 |
|
- train_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adafactor |
|
- lr_scheduler_type: linear |
|
- num_epochs: 2.0 |
|
- mixed_precision_training: Native AMP fp16 |
|
|
|
### Framework versions |
|
|
|
- Transformers 4.30.0.dev0 |
|
- Pytorch 2.0.0+cu117 |
|
- Datasets 2.10.1 |
|
- Tokenizers 0.13.2 |
|
|
