# Units of magnotometer ERFs after exporting to EEGLAB

**URL:** <https://mne.discourse.group/t/units-of-magnotometer-erfs-after-exporting-to-eeglab/11713>\
**Category:** Support & Discussions\
**Tags:** eeglab\
**Created:** [February 25, 2026, 10:47pm UTC](https://mne.discourse.group/t/units-of-magnotometer-erfs-after-exporting-to-eeglab/11713 "2026-02-25T22:47:28Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Mark\_Lavelle](https://yyz2.discourse-cdn.com/free1/user_avatar/mne.discourse.group/mark_lavelle/32/4653_2.png) [@Mark\_Lavelle](https://mne.discourse.group/u/Mark_Lavelle)\
**Post date:** [February 25, 2026, 10:47pm UTC](https://mne.discourse.group/t/units-of-magnotometer-erfs-after-exporting-to-eeglab/11713/1 "2026-02-25T22:47:28Z")

</div>

I’m exporting MEG data to EEGLAB using

`mne.export.export_epochs(path, epochs, 'EEGLAB', overwrite=True)`

I’m confused by the data units contained in the output, once I open it and start analyzing it in EEGLAB.

For my analysis purposes I’m focusing on magnetometers. In the topographic map of the average error-minus-correct ERF (response locked), the autogenerated scale is `± 1.5 E-7` (see below). Units are not specified anywhere in the EEG struct as far as I can tell. The corresponding plots from MNE are displayed on the scale of `± 240 fT`. (Note the EEGLAB topoplot is based on a subset of the participants as the MNE topoplot, which accounts for the slight differences).

If the EEGLAB plot was in `T` units, I would expect the scale to be on the order of `E-13`. Why is this not the case?

 ![image](https://global.discourse-cdn.com/free1/uploads/mne/original/2X/b/bae467baad6b89902ef4b10d9b3374d54fe66aee.jpeg)

 ![image](https://global.discourse-cdn.com/free1/uploads/mne/original/2X/2/2d4af5029626227e931a157fdf4b880780458354.jpeg)

```python
Platform: macOS-12.7.5-x86_64-i386-64bit
Python: 3.10.4 | packaged by conda-forge | (main, Mar 24 2022, 17:43:32) \[Clang 12.0.1 \]
mne: 1.0.0

```