Quazim0t0/Phi4.Turn.R1Distill-Lora4

Phi4 Turn R1Distill LoRA Adapters

Overview

Hey! These LoRA adapters are trained using different reasoning datasets that utilize Thought and Solution for reasoning responses.
I hope these help jumpstart your project! These adapters have been trained on an A800 GPU and should provide a solid base for fine-tuning or merging.

Everything on my page is left public for Open Source use.

Available LoRA Adapters

Here are the links to the available adapters as of January 30, 2025:

• Phi4.Turn.R1Distill-Lora1
• Phi4.Turn.R1Distill-Lora2
• Phi4.Turn.R1Distill-Lora3
• Phi4.Turn.R1Distill-Lora4
• Phi4.Turn.R1Distill-Lora5
• Phi4.Turn.R1Distill-Lora6
• Phi4.Turn.R1Distill-Lora7
• Phi4.Turn.R1Distill-Lora8

Usage

These adapters can be loaded and used with peft and transformers. Here’s a quick example:

from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

base_model = "microsoft/Phi-4"
lora_adapter = "Quazim0t0/Phi4.Turn.R1Distill-Lora1"

tokenizer = AutoTokenizer.from_pretrained(base_model)
model = AutoModelForCausalLM.from_pretrained(base_model)
model = PeftModel.from_pretrained(model, lora_adapter)

model.eval()