mradermacher/loggenix-moe-0.3B-A0.1B-e3-lr7e5-b16-4090-GGUF

Same weights, converted vs kshitijthakkar/loggenix-moe-0.3B-A0.1B-e3-lr7e5-b16-4090. Same weights stored in another dtype (BF16 → F16/F32), plus 36 new tensors.

Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/49, MLP 0/12, Attention 0/48

On the Hub · Downloads on Model Pulse