[Pooling] Honor max_embed_len for chunked embeddings (#55551)
Signed-off-by: Taneem Ibrahim <taneem.ibrahim@gmail.com>
This commit is contained in:
@@ -179,7 +179,7 @@ class EmbeddingTokenizeParamsMixin(PoolingTokenizeParamsMixin):
|
|||||||
pooler_config = model_config.pooler_config
|
pooler_config = model_config.pooler_config
|
||||||
if pooler_config is not None:
|
if pooler_config is not None:
|
||||||
if pooler_config.enable_chunked_processing:
|
if pooler_config.enable_chunked_processing:
|
||||||
max_total_tokens = None
|
max_total_tokens = pooler_config.max_embed_len
|
||||||
else:
|
else:
|
||||||
max_embed_len = pooler_config.max_embed_len or default_max_total_tokens
|
max_embed_len = pooler_config.max_embed_len or default_max_total_tokens
|
||||||
max_output_tokens = default_max_total_tokens - max_embed_len
|
max_output_tokens = default_max_total_tokens - max_embed_len
|
||||||
|
|||||||
Reference in New Issue
Block a user