DuoNeural/LFM2.5-8B-A1B-Hermes-Agentic-Coder-Abliterated-v2
Partial fine-tune vs DuoNeural/LFM2.5-8B-A1B-Abliterated. Partial change: 5% of the weights changed (attention, mlp, other), in 24 of 24 layers.
Tensors changed: MLP 6/2162, Embeddings 0/2, other 36/54, Norms 0/60, Attention 24/24
3 derivatives checked
- mradermacher/LFM2.5-8B-A1B-Hermes-Agentic-Coder-Abliterated-v2-i1-GGUF: Quantization
- mradermacher/LFM2.5-8B-A1B-Hermes-Agentic-Coder-Abliterated-v2-GGUF: Same weights, converted
- DuoNeural/LFM2.5-8B-A1B-Hermes-Agentic-Coder-Abliterated-v2-GGUF: Same weights, converted