diff --git a/README.md b/README.md
index e2568aed..9c983f9f 100644
--- a/README.md
+++ b/README.md
@@ -13,7 +13,7 @@ NequIP is an open-source code for building E(3)-equivariant interatomic potentia
 NequIP requires:
 
 * Python >= 3.7
-* PyTorch == 1.10.* or 1.13.*. PyTorch can be installed following the [instructions from their documentation](https://pytorch.org/get-started/locally/). Note that neither `torchvision` nor `torchaudio`, included in the default install command, are needed for NequIP.
+* PyTorch == `1.11.*` or `1.13.*` or later (do **not** use `1.12`). (Some users have observed silent issues with PyTorch 2+, as reported in #311. Please report any similar issues you encounter.) PyTorch can be installed following the [instructions from their documentation](https://pytorch.org/get-started/locally/). Note that neither `torchvision` nor `torchaudio`, included in the default install command, are needed for NequIP.
 
 **You must install PyTorch before installing NequIP, however it is not marked as a dependency of `nequip` to prevent `pip` from trying to overwrite your PyTorch installation.**
 
@@ -22,14 +22,9 @@ To install:
 * We use [Weights&Biases](https://wandb.ai) (or TensorBoard) to keep track of experiments. This is not a strict requirement — you can use our package without it — but it may make your life easier. If you want to use it, create an account [here](https://wandb.ai) and install the Python package:
 
   ```
-  pip install wandb # tensorboard
+  pip install wandb
   ```
 
-  * for TensorBoard users
-    * On your local computer, build an ssh tunnel to your compute node by `ssh -L 6006:127.0.0.1:6006 username@ip`
-    * On the compute node, go to the `{root}` folder specify in the config file, and run `tensorboard --logdir tb_summary`
-    * Use your local computer browser to log on `http://localhost:6006`
-
 * Install NequIP
 
   NequIP can be installed from PyPI:
@@ -142,7 +137,7 @@ For installation instructions, please see the [`pair_nequip` repository](https:/
 `nequip` is a modular framework and extension packages can provide new model components, architectures, etc. The main extension package(s) currently available are:
  - [Allegro](https://github.com/mir-group/allegro): implements the highly parallelizable Allegro model architecture.
 
-Details on writing and using plugins can be found in the [Allegro tutorial](https://colab.research.google.com/drive/1yq2UwnET4loJYg_Fptt9kpklVaZvoHnq).
+Details on writing and using plugins can be found in the [Allegro tutorial](https://colab.research.google.com/drive/1yq2UwnET4loJYg_Fptt9kpklVaZvoHnq) and in [`nequip-example-extension`](https://github.com/mir-group/nequip-example-extension/).
 
 ## References & citing