diff --git a/examples/train_model.py b/examples/train_model.py
index e115c364281..b926bce4bb2 100644
--- a/examples/train_model.py
+++ b/examples/train_model.py
@@ -145,8 +145,7 @@ def validate(self):
         valid_report, self.valid_world = run_eval(
             self.agent, opt, 'valid', opt['validation_max_exs'],
             valid_world=self.valid_world)
-        if valid_report[opt['validation_metric']] > self.best_valid or \
-                valid_report[opt['validation_metric']] == 1:
+        if valid_report[opt['validation_metric']] > self.best_valid:
             self.best_valid = valid_report[opt['validation_metric']]
             self.impatience = 0
             print('[ new best {}: {} ]'.format(
diff --git a/parlai/agents/seq2seq/seq2seq.py b/parlai/agents/seq2seq/seq2seq.py
index 868e4b709bb..533688578d4 100644
--- a/parlai/agents/seq2seq/seq2seq.py
+++ b/parlai/agents/seq2seq/seq2seq.py
@@ -341,7 +341,7 @@ def predict(self, xs, ys=None, cands=None, valid_cands=None, lm=False):
             self.zero_grad()
             loss = 0
             predictions, scores, _ = self.model(xs, ys)
-            for i in range(ys.size(1)):
+            for i in range(scores.size(1)):
                 # sum loss per-token
                 score = scores.select(1, i)
                 y = ys.select(1, i)