Views
No views yet
⚠️ This is a preview release — expect improvements in future versions.
| Benchmark | Accuracy | Metric |
|---|---|---|
| BoolQ | 57.5% | acc |
| PIQA | 57.8% | acc_norm |
| WinoGrande | 53.3% | acc |
| ARC-Easy | 39.0% | acc_norm |
| HellaSwag | 28.1% | acc_norm |
| Lambada | 11.7% | acc |
1from transformers import AutoModelForCausalLM, AutoTokenizer
2
3tokenizer = AutoTokenizer.from_pretrained("yourorg/sable-mini-30m-preview")
4model = AutoModelForCausalLM.from_pretrained("yourorg/sable-mini-30m-preview")
5
6inputs = tokenizer("The future of tiny models is", return_tensors="pt")
7outputs = model.generate(**inputs, max_new_tokens=50)
8print(tokenizer.decode(outputs[0], skip_special_tokens=True))| Property | Value |
|---|---|
| Parameters | 30M |
| Architecture | Llama-style decoder |
| Training tokens | ~3–4B |
| Vocab size | 24k |