mradermacher/loggenix-moe-0.12B-A0.08B-e5-lr5e4-b4-3060-GGUF

Same weights, converted vs kshitijthakkar/loggenix-moe-0.12B-A0.08B-e5-lr5e4-b4-3060. Same weights stored in another dtype (BF16 → F16/F32), plus 24 new tensors.

Tensors changed: Embeddings 0/1, Norms 0/33, MLP 0/8, Attention 0/32

On the Hub · Downloads on Model Pulse