qwen2.5-3b

Instruction Residuals

This repository contains instruction residuals (delta weights) computed as the parameter-wise difference between Qwen/Qwen2.5-3B-Instruct and Qwen/Qwen2.5-3B.

Apply these residuals to the base model to reconstruct the instruction-tuned weights without retraining.

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer
from residuals import Residuals

base = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-3B")
tok = AutoTokenizer.from_pretrained("Qwen/Qwen2.5-3B")

res = Residuals.from_pretrained("residuals/qwen2.5-3b")
res.apply(base, base_tokenizer=tok)

Provenance

• Created at: 2025-10-25T16:05:19.565132+00:00
• DType: float32
• Parameters: 435
• Shapes hash: 0270b462a2c7458fba25bc5be97a2b6d6e03eb11e573c8f621387231afa4791d
• Names hash: 05a2495a9fd8df03da07f3407419e4bc7e10cd0baaf571b25d9e1fb0e6812a46
• Base model: Qwen/Qwen2.5-3B
• Instruction model: Qwen/Qwen2.5-3B-Instruct

Files

• model.safetensors: Serialized residual tensors (safetensors format).
• (optional) model.safetensors.index.json + shard files model-00001-of-000N.safetensors, ... for multi-part weights.
• config.json: Residuals metadata and provenance.
• tokenizer files: Saved tokenizer for compatibility.

About this format

These are additive residuals (task vectors). Applying them to the base model's parameters reconstructs the instruction-tuned model.

Tools

Generated with the residuals Python package. Install via: pip install residuals.

• PyPI: https://pypi.org/project/residuals/
• Source: https://github.com/omarish/residuals