mradermacher/NeuralStar-7b-Lazy-GGUF

Same weights, converted vs OmnicromsBrain/NeuralStar-7b-Lazy. Same weights stored in another dtype (BF16 → F16/F32).

Tensors changed: Attention 0/128, MLP 0/96, Norms 0/65, Embeddings 0/1, lm_head 0/1

On the Hub · Downloads on Model Pulse