Fix stuff and make tests pass

lm_eval/models/gpt2.py

@@ -60,12 +60,16 @@ class GPT2LM(LM):
                 greedy_tokens = logits.argmax(dim=-1)
                 max_equal = (greedy_tokens == cont_toks).all()
 
-                logits = torch.gather(logits, 2, cont_toks.unsqueeze(-1)).squeeze(-1) # [batch, seq]
+                last_token_slice = logits[:, -1, :].squeeze(0).tolist()
 
+                logits = torch.gather(logits, 2, cont_toks.unsqueeze(-1)).squeeze(-1) # [batch, seq]
 
-                res.append((float(logits.sum()), bool(max_equal)))
+                res.append((float(logits[:, :-1].sum() if logits.shape[-1] > 1 else 0), last_token_slice, bool(max_equal)))
 
-        return reord.get_original(res)
+        # optimization: if two requests have everything the same except the last token, use 
+        # last token distribution to save compute
+        lasttoks = [self.tokenizer.encode(x[1])[-1] for x in requests]
+        return [(l + lts[lasttok], m) for (l, lts, m), lasttok in zip(reord.get_original(res), lasttoks)]
     
     def greedy_until(self, requests):
         # TODO: implement fully general `until` that handles untils that are 

lm_eval/tasks/glue.py

@@ -334,7 +334,7 @@ class MRPC(HFTask):
         return True
 
     def has_test_docs(self):
-        return True
+        return False
 
     def fewshot_description(self):
         return "Indicate if both sentences mean the same thing."
@@ -386,7 +386,7 @@ class QQP(HFTask):
         return True
 
     def has_test_docs(self):
-        return True
+        return False
 
     def fewshot_description(self):
         return "Indicate if both questions ask the same thing."

tests/test_evaluator.py

@@ -29,4 +29,4 @@ def test_evaluator(taskname, Task):
         
 
     lm.loglikelihood = ll_fn
-    evaluator.evaluate(lm, task_dict, False, 0, 10)
+    evaluator.evaluate(lm, task_dict, False, 0, 3)
\ No newline at end of file