KO | EN

GitLite

Minimalist Open Source Repository Search Engine

gitlite — search

#dsh #dsh-plugin #ai-agents #developer-tools #codex #claude-code #mcp #rust #deepseek-harness #typescript #ai-agent #cli

drive_suppress_brains

★ 29 Open GitHub ↗

No description available.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

## Driving and suppressing the human language network using large language models

This repository contains code and data accompanying:

Tuckute, G., Sathe, A., Srikant, S., Taliaferro, M., Wang, M., Schrimpf, M., Kay, K., Fedorenko, E.: _Driving and suppressing the human language network using large language models_. Nature Human Behavior (2024). https://doi.org/10.1038/s41562-023-01783-7

## Environment
The environment is a Python 3.8.11 environment that makes heavy use of [pandas](https://pandas.pydata.org/), [scikit-learn](https://scikit-learn.org/stable/), [HuggingFace Transformers](https://huggingface.co/docs/transformers/index), and [matplotlib](https://matplotlib.org/). To use the exact Python environment used in the paper, install it as:

```
conda env create -f env/env_drive-suppress-brains.yml
```

## Repository organization
 * [data](./data)
 * [data_SI](./data_SI)
 * [env](./env)
 * [model-actv](./model-actv)
   * [gpt2-xl](./model-actv/gpt2-xl)
     * [last-tok](./model-actv/gpt2-xl/last-tok)
 * [regr-weights](./regr-weights)
   * [fit_mapping](./regr-weights/fit_mapping)
     * [last-tok](./regr-weights/fit_mapping/last-tok)
 * [results](./results)
 * [src](./src)
   * [plot_data](./src/plot_data)
   * [run_analyses](./src/run_analyses)
   * [statistics](./src/statistics)
 * [README.md](./README.md)

To populate the `data`, `data_SI`, `model-actv`, and `regr-weights` folders, please see the "Downloading data" section below.

The `data` folder contains a csv file with the event-related data (_brain-lang-data_participant_20230728.csv_; main experiment), a csv file for the blocked experiment (_brain-lang-blocked-data_participant_20230728.csv_), a csv file with the noise ceilings computed based on the event-related data (_NC-allroi-data.csv_), and finally a file with the associated column name descriptions (_column_name_descriptions.csv_). These files are used to run the [Figure_2.ipynb](https://github.com/gretatuckute/drive_suppress_brains/blob/main/src/plot_data/Figure2.ipynb), [Figure_3.ipynb](https://github.com/gretatuckute/drive_suppress_brains/blob/main/src/plot_data/Figure3.ipynb), [Figure_4.ipynb](https://github.com/gretatuckute/drive_suppress_brains/blob/main/src/plot_data/Figure4.ipynb), and [Figure_5.ipynb](https://github.com/gretatuckute/drive_suppress_brains/blob/main/src/plot_data/Figure5.ipynb) notebooks.

The `data_SI` folder contains csv files used to run the [SI_Figures.ipynb](https://github.com/gretatuckute/drive_suppress_brains/blob/main/src/plot_data/SI_Figures.ipynb).

The `env` folder contains the conda yml file _env_drive-suppress-brains.yml_.

The `model-actv` folder contains pre-computed model activations for GPT2-XL (last-token representation). The file _beta-control-neural-T_actv.pkl_ contains the activations for the _baseline_ set in a Pandas dataframe. The rows correspond to sentences, and the columns are multi-indexed according to layer and unit. The first level is layer (49 layers in GPT2-XL) and the second level is unit (1600 units in each representation vector in GPT2-XL). The file _beta-control-neural-T_stim.pkl_ contains the corresponding stimuli metadata in a Pandas dataframe. The two files are row-indexed using the same identifiers. The files _beta-control-neural-D_actv.pkl_ and _beta-control-neural-D_stim.pkl_ contain the activations for the _baseline_ set along with the _drive_/_suppress_ activations (derived via the main search approach).

The `regr-weights` folder contains the encoding model regression weights in the `fit_mapping` subfolder with an additional subfolder according to the parameters that were used to fit the encoding model.

The `results` folder is the default folder for storing outputs from `src/run_analyses`.

The `src` folder contains all code in the following subfolders: 
- `plot_data` contains a notebook that reproduces each of the main figures, as well as a notebook for the SI figures.
-  `run_analyses` contains code to run all main analyses in the paper.
-  `statistics` contains linear mixed effect (LME) statistics (in R).

## Downloading data
To download data used in the paper, download files in the associated [OSF repository](https://osf.io/ru38b). If you are interested in the core analyses, just download the `data` folder (~28 MB). The folder needs to be untarred.

The `data` folder contains a csv file with the event-related data (_brain-lang-data_participant_20230728.csv_; main experiment). This file contains brain responses for the left hemisphere (LH) language regions for n=10 participants (n=5 _train_ participants, n=5 _evaluation_ participants) along with various metadata and behavioral data for each sentence (n=10 linguistic properties). The `data` folder also contains a csv file with brain responses for the blocked experiment (_brain-lang-blocked-data_participant_20230728.csv_, n=4 _evaluation_ participants). The folder also contains the noise ceilings computed based on the event-related data on n=5 _train_ participants (_NC-allroi-data.csv_). Finally, the file _column_name_descriptions.csv_ contains descriptions of the content of the columns in these csv files.

The additional folders besides `data` are: `data_SI` (~300 MB), `model-actv` (~1.5 GB), and `regr-weights` (~80 KB). Organize the folders according to the Repository organization outlined above.

## Analyzing data and generating plots
All code is in `src`.

- The `src/plot_data` folder contains Jupyter Notebooks that analyze and generate plots for the main results in the paper. 
- The `src/run_analyses` folder contains Python scripts for running analyses. The two main scripts are: 
	1. [/src/run_analyses/fit_mapping.py](https://github.com/gretatuckute/drive_suppress_brains/blob/main/src/run_analyses/fit_mapping.py) fits an encoding model from features from a source model (in this case, GPT2-XL, cached in `model-actv/gpt2-xl`) to the participant-averaged brain data. The script will store outputs in `results` and the fitted regression weights in `regr-weights`.
	2. [/src/run_analyses/use_mapping_external.py](https://github.com/gretatuckute/drive_suppress_brains/blob/main/src/run_analyses/use_mapping_external.py) loads the regression weights from the encoding model and predicts each sentence in the supplied stimulus set.
- The `src/statistics` folder contains R code to run LME models.

## Citation
If you use this repository or data, please cite:

```
@article{Tuckute2024,
  title = {Driving and suppressing the human language network using large language models},
  author = {Tuckute, Greta and Sathe, Aalok and Srikant, Shashank and Taliaferro, Maya and Wang, Mingye and Schrimpf, Martin and Kay, Kendrick and Fedorenko, Evelina},
  journal = {Nature Human Behaviour},
  year = {2024},
  date = {2024/01/03},
  abstract = {Transformer models such as GPT generate human-like language and are predictive of human brain responses to language. Here, using functional-MRI-measured brain responses to 1,000 diverse sentences, we first show that a GPT-based encoding model can predict the magnitude of the brain response associated with each sentence. We then use the model to identify new sentences that are predicted to drive or suppress responses in the human language network. We show that these model-selected novel sentences indeed strongly drive and suppress the activity of human language areas in new individuals. A systematic analysis of the model-selected sentences reveals that surprisal and well-formedness of linguistic input are key determinants of response strength in the language network. These results establish the ability of neural network models to not only mimic human language but also non-invasively control neural activity in higher-level cortical areas, such as the language network.},
  doi = {10.1038/s41562-023-01783-7},
  url = {https://doi.org/10.1038/s41562-023-01783-7},
  issn = {2397-3374}
}
```

drive_suppress_brains

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README