|
|
--- |
|
|
library_name: transformers |
|
|
license: mit |
|
|
datasets: |
|
|
- AhmedSSabir/Japanese-wiki-dump-sentence-dataset |
|
|
language: |
|
|
- en |
|
|
- ja |
|
|
base_model: |
|
|
- Qwen/Qwen3-0.6B |
|
|
pipeline_tag: text-generation |
|
|
--- |
|
|
|
|
|
# GEN-1 |
|
|
 |
|
|
## About GEN-1 |
|
|
Model with about 600 million parameters. |
|
|
Japanese-specific SLM. |
|
|
|
|
|
## How to use |
|
|
```py |
|
|
from transformers import pipeline |
|
|
|
|
|
pipe = pipeline("text-generation", model="Ichi075/GEN-1") |
|
|
messages = [ |
|
|
{"role": "user", "content": "Who are you?"}, |
|
|
] |
|
|
pipe(messages) |
|
|
``` |
|
|
```py |
|
|
from transformers import AutoTokenizer, AutoModelForCausalLM |
|
|
|
|
|
tokenizer = AutoTokenizer.from_pretrained("Ichi075/GEN-1") |
|
|
model = AutoModelForCausalLM.from_pretrained("Ichi075/GEN-1") |
|
|
``` |
|
|
|
