Fixed horizon_length for PPLM (#13886)

* fixed horizon_length

* fixed horizon_length

* fix style

examples/research_projects/pplm/run_pplm.py

@@ -181,7 +181,14 @@ def perturb_past(
             for _ in range(horizon_length):
                 inputs_embeds = torch.matmul(curr_probs, wte.weight.data)
                 lm_output = model(past_key_values=curr_unpert_past, inputs_embeds=inputs_embeds)
-                curr_unpert_past, curr_all_hidden = lm_output["past_key_values"], lm_output["hidden_states"]
+                curr_all_logits, curr_unpert_past, curr_all_hidden = (
+                    lm_output["logits"],
+                    lm_output["past_key_values"],
+                    lm_output["hidden_states"],
+                )
+                curr_logits = curr_all_logits[:, -1, :]
+                curr_probs = nn.functional.softmax(curr_logits, dim=-1)
+                curr_probs = torch.unsqueeze(curr_probs, dim=1)
                 curr_hidden = curr_all_hidden[-1]
                 new_accumulated_hidden = new_accumulated_hidden + torch.sum(curr_hidden, dim=1)