project-nils โ project-nils-run-b
Training checkpoints for project-nils.
- Base architecture: Qwen3.5-4B
- RL stage, variant B. Contents:
final/โ full checkpoint (model.safetensors + DeepSpeed ZeRO shards). - Weights are
model.safetensorsper checkpoint;global_step*/folders hold DeepSpeed ZeRO optimizer shards (only needed to resume training). Usezero_to_fp32.pyto reconstruct fp32 weights from shards if needed.
No further details are provided.
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐ Ask for provider support