Lattice Pulse 2 (8B) — LoRA adapter

QLoRA fine-tune of Qwen/Qwen3-8B for Lattice Systems.

This repo is the adapter only (~170MB). Inference still loads the base model, then applies this LoRA:

from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

base = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-8B", torch_dtype="auto", device_map="auto")
model = PeftModel.from_pretrained(base, "oli-mebberson/lattice-pulse-2-8b")
tok = AutoTokenizer.from_pretrained("oli-mebberson/lattice-pulse-2-8b")

Or locally: python -m pulse.chat_pulse2 --adapter /path/to/checkpoint-400

Downloads last month: -

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for oli-mebberson/lattice-pulse-2-8b

Base model

Qwen/Qwen3-8B-Base

Finetuned

Qwen/Qwen3-8B

Adapter

(1940)

this model