Hastagaras/Halu-8B-Llama3-Blackroot
Adapter vs Blackroot/Llama-3-8B-Abomination-LORA. Blackroot/Llama-3-8B-Abomination-LORA is a LoRA adapter, not full weights: rank 64, on down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj in 32 layers (224 adapted matrices). It modifies Hastagaras/Halu-8B-Llama3-Blackroot at load time.
8 derivatives checked
- bartowski/Halu-8B-Llama3-Blackroot-GGUF: Same weights, converted
- mradermacher/Halu-8B-Llama3-Blackroot-GGUF: Same weights, converted
- featherless-ai-quants/Hastagaras-Halu-8B-Llama3-Blackroot-GGUF: Quantization
- mradermacher/Halu-8B-Llama3-Blackroot-i1-GGUF: Quantization
- QuantFactory/Halu-8B-Llama3-Blackroot-GGUF: Quantization
- Casual-Autopsy/Halu-L3-Stheno-BlackOasis-8B: Full fine-tune
- Cas-Archive/L3-Penumbral-Mind-RP-8B: Full fine-tune
- banelingz/black_stheno_8B_v1: Full fine-tune