nbeerbower/llama-3-stella-8B
Full fine-tune vs bunnycore/Cognitron-8B. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: lm_head 1/1, Embeddings 1/1, Norms 65/65, MLP 96/96, Attention 128/128
3 derivatives checked
- featherless-ai-quants/nbeerbower-llama-3-stella-8B-GGUF: Quantization
- mradermacher/llama-3-stella-8B-GGUF: Same weights, converted
- nbeerbower/llama-3-stella-truthy-dpo-8B: LoRA-style