Generate: starcoder 🤜 🤛 assisted generation (#23182)

* starcoder has joined the chat * indexing that works for all
huggingface · May 8, 2023 · bbfb9fc · bbfb9fc
1 parent dbc1226
commit bbfb9fc
Show file tree

Hide file tree

Showing 2 changed files with 14 additions and 4 deletions.
diff --git a/src/transformers/generation/utils.py b/src/transformers/generation/utils.py
@@ -4221,6 +4221,9 @@ def assisted_decoding(
         # keep track of which sequences are already finished
         unfinished_sequences = input_ids.new(input_ids.shape[0]).fill_(1)
 
+        # other auxiliary variables
+        max_len = stopping_criteria[0].max_length
+
         this_peer_finished = False  # used by synced_gpus only
         while True:
             if synced_gpus:
@@ -4235,7 +4238,7 @@ def assisted_decoding(
 
             # Assistant: main logic start
             cur_len = input_ids.shape[-1]
-            max_len = stopping_criteria[0].max_length
+            assistant_kv_indexing = 0 if "bloom" not in assistant_model.__class__.__name__.lower() else 1
 
             #  1. Forecast next N tokens using the assistant model. This `for` block can be replaced with a
             # `.generate()` call if we decide to add `past_key_values` as a possible output of generate, as we
@@ -4244,7 +4247,7 @@ def assisted_decoding(
             for _ in range(int(assistant_model.max_assistant_tokens)):
                 # 1.1. use the assistant model to obtain the next candidate logits
                 if "assistant_past_key_values" in model_kwargs:
-                    prev_seq_len = model_kwargs["assistant_past_key_values"][0][0].shape[2]
+                    prev_seq_len = model_kwargs["assistant_past_key_values"][0][assistant_kv_indexing].shape[-2]
                     # `new_token_len` can be 1 or 2 (next token in assistant + last token picked by the larger model)
                     new_token_len = candidate_input_ids.shape[1] - prev_seq_len
                     assist_inputs = candidate_input_ids[:, -new_token_len:]
@@ -4505,6 +4508,13 @@ def _crop_past_key_values(model, past_key_values, maximum_length):
                 )
             )
         past_key_values = tuple(new_past)
+    elif "gptbigcode" in model.__class__.__name__.lower():  # gptbigcode is too
+        if model.config.multi_query:
+            for idx in range(len(past_key_values)):
+                past_key_values[idx] = past_key_values[idx][:, :maximum_length, :]
+        else:
+            for idx in range(len(past_key_values)):
+                past_key_values[idx] = past_key_values[idx][:, :, :maximum_length, :]
     else:
         for idx in range(len(past_key_values)):
             new_past.append(

diff --git a/tests/generation/test_utils.py b/tests/generation/test_utils.py
@@ -1473,7 +1473,7 @@ def test_assisted_decoding_matches_greedy_search(self):
             # may fix in the future: the following models fail with assisted decoding, and need model-specific fixes
             if any(
                 model_name in model_class.__name__.lower()
-                for model_name in ["bigbirdpegasus", "gptbigcode", "led", "mega", "speech2text", "git", "prophetnet"]
+                for model_name in ["bigbirdpegasus", "led", "mega", "speech2text", "git", "prophetnet"]
             ):
                 return
 
@@ -1529,7 +1529,7 @@ def test_assisted_decoding_sample(self):
             # may fix in the future: the following models fail with assisted decoding, and need model-specific fixes
             if any(
                 model_name in model_class.__name__.lower()
-                for model_name in ["bigbirdpegasus", "gptbigcode", "led", "mega", "speech2text", "git", "prophetnet"]
+                for model_name in ["bigbirdpegasus", "led", "mega", "speech2text", "git", "prophetnet"]
             ):
                 return