SentientAGI/Dobby-Unhinged-Llama-3.3-70B
Full fine-tune vs meta-llama/Llama-3.3-70B-Instruct. Fine-tune: 97% of the weights changed, across 80 of 80 layers.
Tensors changed: Embeddings 0/1, Norms 9/161, MLP 240/240, Attention 320/320, lm_head 0/1
7 derivatives checked
- bartowski/SentientAGI_Dobby-Unhinged-Llama-3.3-70B-GGUF: Quantization
- mradermacher/Dobby-Unhinged-Llama-3.3-70B-i1-GGUF: Quantization
- mradermacher/Dobby-Unhinged-Llama-3.3-70B-GGUF: Quantization
- TareksLab/Persona-V4-70B: Full fine-tune
- TareksLab/Persona-V5-70B: Full fine-tune
- TareksLab/Persona-V3-70B: Full fine-tune
- TareksLab/Doppleganger-V10-LLaMa-70B: Full fine-tune