Nexesenex/Llama_3.1_8b_Smarteaz_0.1b
Full fine-tune vs allenai/Llama-3.1-Tulu-3.1-8B. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Norms 65/65, MLP 96/96, Attention 128/128
5 derivatives checked
- mradermacher/Llama_3.x_8b_Smarteaz_0.1b-i1-GGUF: Full fine-tune
- mradermacher/Llama_3.1_8b_Smarteaz_0.1b-i1-GGUF: Quantization
- mradermacher/Llama_3.x_8b_Smarteaz_0.1b-GGUF: Partial fine-tune
- mradermacher/Llama_3.1_8b_Smarteaz_0.1b-GGUF: Same weights, converted
- Nexesenex/Llama_3.1_8b_DeepDive_3_Prev_v1.0: Full fine-tune