[Lora] Support long context lora (#4787)
Currently we need to call rotary embedding kernel for each LoRA, which makes it hard to serve multiple long context length LoRA. Add batched rotary embedding kernel and pipe it through. It replaces the rotary embedding layer to the one that is aware of multiple cos-sin-cache per scaling factors. Follow up of https://github.com/vllm-project/vllm/pull/3095/files
This commit is contained in:
@@ -46,6 +46,8 @@ class ChatGLMConfig(PretrainedConfig):
|
||||
self.kv_channels = kv_channels
|
||||
self.num_attention_heads = num_attention_heads
|
||||
self.seq_length = seq_length
|
||||
# It is to be compatible with long lora.
|
||||
self.max_position_embeddings = seq_length
|
||||
self.hidden_dropout = hidden_dropout
|
||||
self.attention_dropout = attention_dropout
|
||||
self.layernorm_epsilon = layernorm_epsilon
|
||||
|
||||
@@ -34,12 +34,26 @@ class TokenizerGroup(BaseTokenizerGroup):
|
||||
"""Get the maximum input length for the LoRA request."""
|
||||
return self.max_input_length
|
||||
|
||||
def _raise_if_input_too_long(self,
|
||||
encoded_tokens: List[str],
|
||||
lora_request: Optional[LoRARequest] = None):
|
||||
input_length = len(encoded_tokens)
|
||||
if lora_request:
|
||||
max_input_length = (lora_request.long_lora_max_len
|
||||
or self.max_input_length)
|
||||
else:
|
||||
max_input_length = self.max_input_length
|
||||
if max_input_length is not None and input_length > max_input_length:
|
||||
raise ValueError("Input too long.", input_length, max_input_length)
|
||||
|
||||
def encode(self,
|
||||
prompt: str,
|
||||
request_id: Optional[str] = None,
|
||||
lora_request: Optional[LoRARequest] = None) -> List[int]:
|
||||
tokenizer = self.get_lora_tokenizer(lora_request)
|
||||
return tokenizer.encode(prompt)
|
||||
ret = tokenizer.encode(prompt)
|
||||
self._raise_if_input_too_long(ret, lora_request)
|
||||
return ret
|
||||
|
||||
async def encode_async(
|
||||
self,
|
||||
@@ -47,7 +61,9 @@ class TokenizerGroup(BaseTokenizerGroup):
|
||||
request_id: Optional[str] = None,
|
||||
lora_request: Optional[LoRARequest] = None) -> List[int]:
|
||||
tokenizer = await self.get_lora_tokenizer_async(lora_request)
|
||||
return tokenizer.encode(prompt)
|
||||
ret = tokenizer.encode(prompt)
|
||||
self._raise_if_input_too_long(ret, lora_request)
|
||||
return ret
|
||||
|
||||
def get_lora_tokenizer(
|
||||
self,
|
||||
|
||||
Reference in New Issue
Block a user