diff --git a/.gitignore b/.gitignore
index 6db56a4a..ea9c1977 100644
--- a/.gitignore
+++ b/.gitignore
@@ -8,6 +8,7 @@ results
 ./*.json
 client_configs/*.yaml
 old_results
+results_evaluators
 
 # Byte-compiled / optimized / DLL files
 __pycache__/
diff --git a/src/alpaca_eval/decoders/openai.py b/src/alpaca_eval/decoders/openai.py
index df8a0f6c..c3975235 100644
--- a/src/alpaca_eval/decoders/openai.py
+++ b/src/alpaca_eval/decoders/openai.py
@@ -132,10 +132,10 @@ def openai_completions(
 
     prompt_batches = [prompts[batch_id * batch_size : (batch_id + 1) * batch_size] for batch_id in range(n_batches)]
 
-    if isinstance(max_tokens, int):
-        max_tokens = [max_tokens] * n_examples
-
-    inputs = zip(prompt_batches, max_tokens)
+    try:
+        inputs = zip(prompt_batches, max_tokens)
+    except TypeError:
+        inputs = zip(prompt_batches, [max_tokens] * n_batches)
 
     kwargs = dict(model=model_name, **decoding_kwargs)
     kwargs_to_log = {k: v for k, v in kwargs.items() if "api_key" not in k}
@@ -216,7 +216,11 @@ def _openai_completion_helper(
     client = all_clients[curr_client_idx]
 
     # copy shared_kwargs to avoid modifying it
-    kwargs.update(dict(max_tokens=max_tokens, top_p=top_p, temperature=temperature))
+    to_update = dict()
+    for k in ["max_tokens", "top_p", "temperature"]:
+        if locals()[k] is not None:
+            to_update[k] = locals()[k]
+    kwargs.update(to_update)
     curr_kwargs = copy.deepcopy(kwargs)
 
     # ensure no infinite loop
@@ -242,7 +246,7 @@ def _openai_completion_helper(
                         # currently we only use function calls to get a JSON object => return raw text of json
                         choices[i]["text"] = choice.message.function_call.arguments
 
-                    if choice.message.tool_calls is not None:
+                    if choice.message.tool_calls:
                         # currently we only use function calls to get a JSON object => return raw text of json
                         choices[i]["text"] = choice.message.tool_calls[0].function.arguments
 
@@ -273,10 +277,11 @@ def _openai_completion_helper(
                 return choices
 
             else:
-                if "rate limit" in str(e).lower():
+                if "rate " in str(e).lower():
                     logging.warning(f"Hit request rate limit; retrying...")
                 else:
-                    logging.warning(f"Unknown error. \n It's likely a rate limit so we are retrying...")
+                    logging.exception("Unknown error:")
+                    raise e
                 if len(all_clients) > 1:
                     curr_client_idx = random.choice([idx for idx in client_idcs if idx != curr_client_idx])
                     client = all_clients[curr_client_idx]