legendarydragontamer

slugextract-qwen2.5-1.5b-lora

Deploy Dedicated

Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more

Container

Run this model inference with full control and performance in your environment.

Learn more

Get help setting up a custom Dedicated Endpoints.

Talk with our engineer to get a quote for reserved GPU instances with discounts.

README

License: apache-2.0

Training

Base: Qwen/Qwen2.5-1.5B-Instruct
Method: LoRA (r=32, alpha=32) on all attention + MLP projections
Data: 167 (transcript -> extraction JSON) pairs, balanced across 14 sentiment arcs, 100% with non-empty dead_ends
4 epochs, final training loss 0.81
Trained on Modal A10G with transformers + PEFT + bitsandbytes

Companion adapter

The slug's voice comes from a separate adapter: slugvoice-qwen2.5-1.5b-lora. Two adapters, one base model.

Usage

python
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer

base = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-1.5B-Instruct")
model = PeftModel.from_pretrained(base, "legendarydragontamer/slugextract-qwen2.5-1.5b-lora")