[Shardformer] change qwen2 modeling into gradient checkpointing style (#5874)

colossalai/shardformer/modeling/qwen2.py

@@ -168,13 +168,27 @@ class Qwen2PipelineForwards:
         next_decoder_cache = None
 
         start_idx, end_idx = stage_index[0], stage_index[1]
+        num_ckpt_layers = 0
+        if self.gradient_checkpointing and self.training:
+            num_ckpt_layers = end_idx - start_idx
+            # TODO: We can replace `gradient_checkpointing_enable` fn and initialize a gradient_checkpointing (List[bool]) for each layer
+            if shard_config.gradient_checkpoint_config is not None:
+                num_ckpt_layers = shard_config.gradient_checkpoint_config.get_num_ckpt_layers(
+                    stage=stage_manager.stage,
+                    num_stages=stage_manager.num_stages,
+                    num_layers=end_idx - start_idx,
+                    model_chunk_id=(stage_manager.model_chunk_id if stage_manager.is_interleave else 0),
+                    num_model_chunks=stage_manager.num_model_chunks,
+                )
+            assert num_ckpt_layers <= end_idx - start_idx
+
         for idx, decoder_layer in enumerate(self.layers[start_idx:end_idx], start=start_idx):
             if output_hidden_states:
                 all_hidden_states += (hidden_states,)
 
             past_key_value = past_key_values[idx] if past_key_values is not None else None
 
-            if self.gradient_checkpointing and self.training:
+            if idx - start_idx < num_ckpt_layers:
                 layer_outputs = self._gradient_checkpointing_func(
                     decoder_layer.__call__,
                     hidden_states,
@@ -198,7 +212,6 @@ class Qwen2PipelineForwards:
 
             if use_cache:
                 next_decoder_cache += (layer_outputs[2 if output_attentions else 1],)
-
             if output_attentions:
                 all_self_attns += (layer_outputs[1],)