|
|
--- |
|
|
tags: |
|
|
- babylm |
|
|
- language-model |
|
|
- coherence |
|
|
license: mit |
|
|
language: |
|
|
- uk |
|
|
--- |
|
|
|
|
|
# babybabellm-mono-ukr |
|
|
|
|
|
This repository contains checkpoints for the **mono-ukr** variant of **BabyBabeLLM**. |
|
|
|
|
|
## Files |
|
|
- `*_15_16.bin` – main model weights |
|
|
- `*_15_16_ema.bin` – EMA smoothed weights |
|
|
- `*_15_16_state_dict.bin` – PyTorch state dict |
|
|
- `pytorch_model.bin` – extracted EMA weights (for AutoModel) |
|
|
- Config + tokenizer files for model loading |
|
|
|
|
|
## Usage |
|
|
|
|
|
```python |
|
|
from transformers import AutoModel, AutoTokenizer |
|
|
|
|
|
repo = "suchirsalhan/babybabellm-mono-ukr" |
|
|
|
|
|
tokenizer = AutoTokenizer.from_pretrained(repo) |
|
|
model = AutoModel.from_pretrained(repo) |
|
|
|
|
|
inputs = tokenizer("Hello world!", return_tensors="pt") |
|
|
outputs = model(**inputs) |
|
|
``` |
|
|
|
|
|
## Notes |
|
|
- These are research checkpoints trained on BabyLM-style data. |
|
|
- Model naming: `mono-ukr` indicates the language/config variant. |
