bartowski/v6-Finch-7B-HF-GGUF
Same weights, converted vs RWKV/v6-Finch-7B-HF. Same weights stored in another dtype (BF16 → F16/F32), plus 448 new tensors.
Tensors changed: other 0/3, Attention 0/160, Norms 0/194, MLP 0/96, Embeddings 0/1
Same weights, converted vs RWKV/v6-Finch-7B-HF. Same weights stored in another dtype (BF16 → F16/F32), plus 448 new tensors.
Tensors changed: other 0/3, Attention 0/160, Norms 0/194, MLP 0/96, Embeddings 0/1