Fix typo in slc.ipynb

docs/source/slc.ipynb

@@ -697,7 +697,7 @@
     "\n",
     "In words, in this setting the ATE tells us how much greater the students' test scores would be across all individuals if we forced everyone to be shown the \"Electric Company\" educational program relative to if we forced everyone to not be shown the \"Electric Company\" educational program. Here, we are interested in the average over the entire population.\n",
     "\n",
-    "To implement this query in ChiRho we extend our `LinearSLC` model by applying two interventions, `do(actions=dict(T=torch.zeros(treatment_shape))` and `do(actions=dict(T=torch.zeros(treatment_shape))`, and then sampling jointly from counterfactual worlds using the `MultiWorldCounterfactual` handler. Recall from the [tutorial](tutorial_i.ipynb) that the `MultiWorldCounterfactual` handler modifies the execution of the causal model to sample jointly from the observational and all counterfactual worlds induced by an intervention.\n",
+    "To implement this query in ChiRho we extend our `LinearSLC` model by applying two interventions, `do(actions=dict(T=torch.zeros(treatment_shape))` and `do(actions=dict(T=torch.ones(treatment_shape))`, and then sampling jointly from counterfactual worlds using the `MultiWorldCounterfactual` handler. Recall from the [tutorial](tutorial_i.ipynb) that the `MultiWorldCounterfactual` handler modifies the execution of the causal model to sample jointly from the observational and all counterfactual worlds induced by an intervention.\n",
     "\n",
     "**Note:** Here we define the queried model in terms of the \"individual treatment effect\" (ITE), i.e. the estimated difference in outcomes for each individual under treated and control conditions. Given estimates of the ITE, we can later perform post-hoc analyses to produce estimates of post-stratified average treatment effects. Specifically, later we'll show average treatment effects conditional on the school grade."
    ]