mradermacher/LLaDA-8B-Instruct-JSON-SFT-GGUF

Same weights, converted vs tarsur909/LLaDA-8B-Instruct-JSON-SFT. Same weights stored in another dtype (F32 → F16/F32).

Tensors changed: Norms 0/65, Attention 0/128, other 0/65, MLP 0/32, Embeddings 0/1

On the Hub · Downloads on Model Pulse