DavidAU/L3-Darkest-Planet-16B-HERETIC-Uncensored-Abliterated
Partial fine-tune (fewer layers) vs DavidAU/L3-Dark-Planet-8B. DavidAU/L3-Dark-Planet-8B matches part of DavidAU/L3-Darkest-Planet-16B-HERETIC-Uncensored-Abliterated (291 of its 615 tensors; the rest of DavidAU/L3-Darkest-Planet-16B-HERETIC-Uncensored-Abliterated has no counterpart). On the shared part: Partial change: 31% of the weights changed (attention, lm_head, mlp, norms), in 31 of 32 layers.
Tensors changed: lm_head 1/1, Embeddings 0/1, Norms 3/65, MLP 33/96, Attention 4/128
2 derivatives checked
- mradermacher/L3-Darkest-Planet-16B-HERETIC-Uncensored-Abliterated-i1-GGUF: Quantization
- mradermacher/L3-Darkest-Planet-16B-HERETIC-Uncensored-Abliterated-GGUF: Quantization
1 other repos hold byte-for-byte identical weight files.