jiwon9703/Phi-4-mini-instruct-ko-hallucination-sft-v0

Partial fine-tune vs unsloth/Phi-4-mini-instruct. Partial change: 84% of the weights changed (attention, mlp, norms), in 32 of 32 layers.

Tensors changed: Embeddings 0/1, Norms 10/65, MLP 64/64, Attention 64/64

1 derivatives checked

• mradermacher/Phi-4-mini-instruct-ko-hallucination-sft-v0-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse