YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
c3-ft110m-v3-visible
Masked diffusion language model (MDLM objective, absorbing mask, SUBS), fine-tuned for the C3 project (committed-confidence calibration in diffusion LMs).
T9 fine-tune (b): as (a), but the rho marker is never masked (visible 8-token prefix), 300M tokens (W&B ft110m-v3-visible).
| parameters | 123.9M (tied embeddings) |
| training tokens | 299,923,840 |
| steps | 55179 |
| weights | ema |
| config hash | 2b848a1955c93291 |
| code commit | 48b91892d4f5200361b28f80f816f38f1c148218 |
| tokenizer | GPT-2 BPE + one [MASK] token (id 50257) |
- Downloads last month
- 2
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support