mergekit-community/nsfw-merge-v4dot1-w-deepseek-ablit
Full fine-tune vs mergekit-community/nsfw_merge_test_v4dot1. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: lm_head 1/1, Embeddings 1/1, Norms 65/65, MLP 96/96, Attention 128/128
2 derivatives checked
- mradermacher/nsfw-merge-v4dot1-w-deepseek-ablit-i1-GGUF: Quantization
- mradermacher/nsfw-merge-v4dot1-w-deepseek-ablit-GGUF: Same weights, converted