shiningdota/Llama-3.2-3B_Instruct_Indonesian_gguf-test

Finetuned Llama-3.2-3B with MBZUAI-Bactrian-X Indonesian datasets only

Copal_id benchmark (before quantized into gguf):

copal_id_standard = Formal Language

copal_id_standard_multishots	Result
0-shot	56
10-shots	58
25-shots	57

copal_id_colloquial = Informal Language and local nuances

copal_id_colloquial_multishots	Result
0-shot	54
10-shots	54
25-shots	52

Using Unsloth.ai for finetuning