v000000/Llama-3.1-8B-Stheno-v3.4-abliterated

Partial fine-tune vs Sao10K/Llama-3.1-8B-Stheno-v3.4. Partial change: 25% of the weights changed (attention, mlp, norms), in 28 of 32 layers.

Tensors changed: Embeddings 0/1, Norms 1/65, MLP 27/96, Attention 27/128, lm_head 0/1

2 derivatives checked

• mradermacher/Llama-3.1-8B-Stheno-v3.4-abliterated-i1-GGUF: Quantization
• mradermacher/Llama-3.1-8B-Stheno-v3.4-abliterated-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse