birgermoell/llama-3-merge-disco-neural-pace
Full fine-tune vs DiscoResearch/Llama3_DiscoLM_German_8b_v0.1_experimental. Fine-tune: 93% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/1, Norms 5/65, MLP 96/96, Attention 128/128, lm_head 0/1
3 derivatives checked
- featherless-ai-quants/birgermoell-llama-3-merge-disco-neural-pace-GGUF: Quantization
- mradermacher/llama-3-merge-disco-neural-pace-i1-GGUF: Quantization
- mradermacher/llama-3-merge-disco-neural-pace-GGUF: Same weights, converted