huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-10 20:04:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

added a tokenizer to HeterogeneousNextTokenChooser

Browse Source 
I'm not super happy with this solution - but it seemed most in line with the current way of passing in extra params into `from_pb` in `concatenate`.
This commit is contained in:
marcusdunn 2023-08-15 15:12:32 -07:00
parent f65d703de4
commit 25c48f5679
1 changed files with 1 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
server/text_generation_server/models/flash_causal_lm.py
View File

@ -634,8 +634,7 @@ class FlashCausalLMBatch(Batch):
next_token_chooser_parameters,
dtype=batches[0].next_token_chooser.dtype,
device=batches[0].next_token_chooser.device,
# todo - determine how to obtain access to a tokenizer here
tokenizer=...
tokenizer=batches[0].next_token_chooser.tokenizer
)
# Needed to avoid dropping blocks when the batches will go out of scope