YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
Celerity 3900m 8K — ad_0.0_cbcore2.6.0
Official Celerity 3900m / 8K attention-dropout baseline checkpoint converted from Cerebras CS format to Hugging Face format.
Training provenance
- Original experiment:
ad_0.0_cbcore2.6.0 - Source checkpoint:
checkpoint_47389.mdl - Attention dropout rate: 0.0
- Attention-dropout schedule: constant / no attention dropout
- Residual dropout rate: 0.0
- Stochastic depth: 0.0
- LayerDrop: 0.0
- Converter commit:
0e3d5d375695293479df9d2a3717f05f71a345b4
Attention dropout is a training-time hyperparameter. For this baseline it was
disabled at 0.0. Hugging Face evaluation is performed with dropout
disabled in model.eval().
This repository uses custom Celerity Hugging Face modeling code and should be
loaded with trust_remote_code=True.
- Downloads last month
- 19
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support