SicariusSicariiStuff/Eximius_Persona_5B
Full fine-tune (fewer layers) vs meta-llama/Llama-3.2-3B-Instruct. meta-llama/Llama-3.2-3B-Instruct matches part of SicariusSicariiStuff/Eximius_Persona_5B (254 of its 453 tensors; the rest of SicariusSicariiStuff/Eximius_Persona_5B has no counterpart). On the shared part: Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 57/57, MLP 84/84, Attention 112/112
4 derivatives checked
- mradermacher/Eximius_Persona_5B-i1-GGUF: Quantization
- mradermacher/Eximius_Persona_5B-GGUF: Same weights, converted
- SicariusSicariiStuff/Eximius_Persona_5B_GGUF: Quantization
- SicariusSicariiStuff/Eximius_Persona_5B_iMatrix: Quantization
1 other repos hold byte-for-byte identical weight files.