ytu-ce-cosmos/Turkish-Llama-8b-v0.1
Full fine-tune vs meta-llama/Meta-Llama-3-8B. Fine-tune: 93% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 0/1, Norms 26/65, MLP 96/96, Attention 128/128, lm_head 1/1
5 derivatives checked
- mradermacher/Turkish-Llama-8b-v0.1-i1-GGUF: Quantization
- QuantFactory/Turkish-Llama-8b-v0.1-GGUF: Quantization
- featherless-ai-quants/ytu-ce-cosmos-Turkish-Llama-8b-v0.1-GGUF: Quantization
- mradermacher/Turkish-Llama-8b-v0.1-GGUF: Same weights, converted
- kayrab/doktor-llama-3-cosmos-8b: LoRA-style