ReadyArt/Forgotten-Safeword-12B-v4.0
LoRA-style vs TheDrummer/UnslopNemo-12B-v3. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj changed, in 40 of 40 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/81, MLP 120/120, Attention 160/160, lm_head 0/1
9 derivatives checked
- mradermacher/Forgotten-Safeword-12B-v4.0-i1-GGUF: Quantization
- mradermacher/Forgotten-Safeword-12B-v4.0-GGUF: Quantization
- Tsuna57/Forgotten-Safeword-12B-v4.0-Q8_0-GGUF: Quantization
- limloop/MN-12B-Faun-RP-RU: Full fine-tune
- ReadyArt/Forgotten-Abomination-12B-v4.0: LoRA-style
- pot99rta/Patricide-12B-Forgottenslop-Mell: Full fine-tune
- yamatazen/ForgottenMaid-12B: Full fine-tune
- Retreatcost/Forgotten-directive-Neon-stock: Full fine-tune
- neighbooo/Wicked-Love-12B: Full fine-tune
2 other repos hold byte-for-byte identical weight files.