realtime-rubric-scoring-v1b

This is a LoRA checkpoint trained with trainingRun ID: 18cfdf27-dda0-4549-b10e-e1dafa624688 on trajectory by castform.com

Training Details

• Base Model: Qwen3.5-4B
• Checkpoint: lora_peft_iter_0000017
• Training run: realtime-rubric-scoring-v1b

Usage

from transformers import AutoModelForCausalLM
from peft import PeftModel

base_model = AutoModelForCausalLM.from_pretrained("Qwen3.5-4B")
model = PeftModel.from_pretrained(base_model, "alwang/realtime-rubric-scoring-v1b")