mradermacher/chohee-gemma-3-4b-response-false-GGUF

Same weights, converted (fewer layers) vs oneandthree/chohee-gemma-3-4b-response-false. mradermacher/chohee-gemma-3-4b-response-false-GGUF matches part of oneandthree/chohee-gemma-3-4b-response-false (444 of its 883 tensors; the rest of oneandthree/chohee-gemma-3-4b-response-false has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).

Tensors changed: Embeddings 0/1, Norms 0/205, MLP 0/102, Attention 0/136

On the Hub · Downloads on Model Pulse