mradermacher/F2LLM-v2-80M-GGUF
Same weights, converted vs codefuse-ai/F2LLM-v2-80M. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/33, MLP 0/24, Attention 0/32
Same weights, converted vs codefuse-ai/F2LLM-v2-80M. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/33, MLP 0/24, Attention 0/32