---
library_name: residuals
base_model: ibm-granite/granite-4.0-micro-base
base_model_relation: adapter
instruct_model: ibm-granite/granite-4.0-micro
pipeline_tag: text-generation
tags:
- residuals
- delta
- task-arithmetic
- finetune
---


# Instruction Residuals

This repository contains instruction residuals (delta weights) computed as the parameter-wise difference between `ibm-granite/granite-4.0-micro` and `ibm-granite/granite-4.0-micro-base`.

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


## Usage

```python
from transformers import AutoModelForCausalLM, AutoTokenizer
from residuals import Residuals

base = AutoModelForCausalLM.from_pretrained("ibm-granite/granite-4.0-micro-base")
tok = AutoTokenizer.from_pretrained("ibm-granite/granite-4.0-micro-base")

res = Residuals.from_pretrained("residuals/granite-4.0-micro")
res.apply(base, base_tokenizer=tok)
```


## Provenance
- **Created at**: 2025-10-25T17:37:57.309504+00:00
- **DType**: float32
- **Parameters**: 323
- **Shapes hash**: 76567e0baeae70e88388dfccd02356eb6217113844059502848307ae05929faf
- **Names hash**: 390466ca330d23f03baa73ba7f44c2df87ff370c3b94fffd4cdd7a6df22f820b
- **Base model**: `ibm-granite/granite-4.0-micro-base`
- **Instruction model**: `ibm-granite/granite-4.0-micro`

## 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