huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-11 12:24:53 +00:00
Code Issues Packages Projects Releases Wiki Activity

"Fixing" double BOS for mistral too.

Browse Source
This commit is contained in:
Nicolas Patry 2024-05-01 18:20:44 +02:00
parent 6073ece4fc
commit d1639a5827
1 changed files with 5 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
server/text_generation_server/models/flash_mistral.py
View File

@ -121,6 +121,11 @@ class FlashMistralBatch(FlashCausalLMBatch):
requests_idx_mapping[r.id] = i
tokenized_input = tokenized_input[-r.truncate :]
if (
tokenized_input[0] == tokenizer.bos_token_id
and tokenized_input[1] == tokenizer.bos_token_id
):
tokenized_input = tokenized_input[1:]
input_length = len(tokenized_input)
input_lengths.append(input_length)