DebateLabKIT/Llama-3.3-Argunaut-1-70B-SFT
Partial fine-tune vs meta-llama/Llama-3.3-70B-Instruct. Partial change: 32% of the weights changed (attention, embeddings, lm_head, mlp), in 71 of 80 layers.
Tensors changed: Embeddings 1/1, Norms 0/161, MLP 72/240, Attention 96/320, lm_head 1/1
3 derivatives checked
- mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF: Quantization
- mradermacher/Llama-3.3-Argunaut-1-70B-SFT-GGUF: Quantization
- DebateLabKIT/Llama-3.3-Argunaut-1-70B-SPIN: Partial fine-tune