diff --git a/doc/source/rllib-concepts.rst b/doc/source/rllib-concepts.rst
index a81029c4e..f629d4ad1 100644
--- a/doc/source/rllib-concepts.rst
+++ b/doc/source/rllib-concepts.rst
@@ -423,8 +423,8 @@ Building Policies in TensorFlow Eager
 
 Policies built with ``build_tf_policy`` (most of the reference algorithms are)
 can be run in eager mode by setting
-the ``"eager": True`` / ``"eager_tracing": True`` config options or
-using ``rllib train --eager [--trace]``.
+the ``"framework": "tf2"`` / ``"eager_tracing": True`` config options or
+using ``rllib train '{"framework": "tf2", "eager_tracing": True}'``.
 This will tell RLlib to execute the model forward pass, action distribution,
 loss, and stats functions in eager mode.
 
diff --git a/doc/source/rllib-toc.rst b/doc/source/rllib-toc.rst
index 9f4a497d9..86a2b3f34 100644
--- a/doc/source/rllib-toc.rst
+++ b/doc/source/rllib-toc.rst
@@ -222,7 +222,7 @@ references in the cluster.
 TensorFlow 2.0
 ~~~~~~~~~~~~~~
 
-RLlib currently runs in ``tf.compat.v1`` mode. This means eager execution is disabled by default, and RLlib imports TF with ``import tensorflow.compat.v1 as tf; tf.disable_v2_behaviour()``. Eager execution can be enabled manually by calling ``tf.enable_eager_execution()`` or setting the ``"eager": True`` trainer config.
+RLlib currently runs in ``tf.compat.v1`` mode. This means eager execution is disabled by default, and RLlib imports TF with ``import tensorflow.compat.v1 as tf; tf.disable_v2_behaviour()``. Eager execution can be enabled manually by calling ``tf.enable_eager_execution()`` or setting the ``"framework": "tf2"`` trainer config.
 
 .. |tensorflow| image:: tensorflow.png
     :class: inline-figure
diff --git a/doc/source/rllib-training.rst b/doc/source/rllib-training.rst
index c88b7ecdd..1bfbc7053 100644
--- a/doc/source/rllib-training.rst
+++ b/doc/source/rllib-training.rst
@@ -14,7 +14,7 @@ You can train a simple DQN trainer with the following command:
 
 .. code-block:: bash
 
-    rllib train --run DQN --env CartPole-v0  # --eager [--trace] for eager execution
+    rllib train --run DQN --env CartPole-v0  # --config '{"framework": "tf2", "eager_tracing": True}' for eager execution
 
 By default, the results will be logged to a subdirectory of ``~/ray_results``.
 This subdirectory will contain a file ``params.json`` which contains the
@@ -906,7 +906,7 @@ Eager Mode
 Policies built with ``build_tf_policy`` (most of the reference algorithms are)
 can be run in eager mode by setting the
 ``"framework": "[tf2|tfe]"`` / ``"eager_tracing": True`` config options or using
-``rllib train --eager [--trace]``.
+``rllib train --config '{"framework": "tf2"}' [--trace]``.
 This will tell RLlib to execute the model forward pass, action distribution,
 loss, and stats functions in eager mode.
 
diff --git a/doc/source/rllib.rst b/doc/source/rllib.rst
index 33a808a04..2f045e0d5 100644
--- a/doc/source/rllib.rst
+++ b/doc/source/rllib.rst
@@ -28,8 +28,8 @@ Then, you can try out training in the following equivalent ways:
 .. code-block:: bash
 
   rllib train --run=PPO --env=CartPole-v0  # -v [-vv] for verbose,
-                                           # --eager [--trace] for eager execution,
-                                           # --torch to use PyTorch
+                                           # --config='{"framework": "tf2", "eager_tracing": True}' for eager,
+                                           # --torch to use PyTorch OR --config='{"framework": "torch"}'
 
 .. code-block:: python