Qubitium (Qubitium) – Community Activity

#4 opened 12 days ago by

Orenguteng

New activity in nvidia/Llama3-ChatQA-1.5-8B 13 days ago

The tokenizer adds a special token '<|im_end|>' to solve the problem of non-stop generation when encountering <|im_end|>.

#16 opened 19 days ago by

zjyhf

New activity in Falconsai/nsfw_image_detection 21 days ago

False positives

#6 opened 2 months ago by

Dolfik

New activity in databricks/dbrx-base about 2 months ago

Tokenizer change has strange order plus format

#21 opened about 2 months ago by

New activity in CohereForAI/c4ai-command-r-v01 about 2 months ago

Fix AutoTokenizerFromPretrained crash with transformer 4.39.3

3

#48 opened about 2 months ago by

Could not locate the tokenization_cohere_fast.py

3

#47 opened about 2 months ago by

falafelfrank

New activity in Xenova/dbrx-instruct-tokenizer about 2 months ago

Full 1-1 compat with dbrx-instruct tokenizer?

#1 opened 2 months ago by

New activity in LnL-AI/dbrx-base-converted-v2 2 months ago

Ready for Testing...

9

#1 opened 2 months ago by

New activity in databricks/dbrx-base 2 months ago

Why is config.vocab_size != tokenizer.vocab_size?

#18 opened 2 months ago by

Train datasets?

#11 opened 2 months ago by

danielpark

New activity in LnL-AI/dbrx-base-converted-v2 2 months ago

Update tiktoken.py

#4 opened 2 months ago by

New activity in LnL-AI/dbrx-base-tokenizer 2 months ago

Upload tokenizer.json

#5 opened 2 months ago by

fix bad commit

#4 opened 2 months ago by

bad commit

#3 opened 2 months ago by

fix tokenizer.json

#2 opened 2 months ago by

remove instruct tokens from json

#1 opened 2 months ago by

New activity in databricks/dbrx-instruct 2 months ago

The fused expert parameters means load_in_4bit doesn't work properly, nor does LoRA

31

#10 opened 2 months ago by

tdrussell

New activity in mistralai/Mistral-7B-Instruct-v0.2 2 months ago

Update config.json to accurately reflect the 32k context window.

#73 opened 2 months ago by

Kearm

New activity in v2ray/dbrx-base-fixed 2 months ago

FlashAttention2 support not working during training

#1 opened 2 months ago by

New activity in LnL-AI/dbrx-base-converted-v2 2 months ago

reduce logging verbosity

#3 opened 2 months ago by

winglian

Fix import typo

#2 opened 2 months ago by

fahadh4ilyas

New activity in fishaudio/speech-lm-v1 5 months ago

Upload config.json

#1 opened 5 months ago by

tianruoyou

New activity in microsoft/phi-2 6 months ago

Feature Request - add support for beam search

#30 opened 6 months ago by

miguelcarv

Nonsense Chinese output! This might be expected since it is a unaligned model?

6

#26 opened 6 months ago by

Yhyu13

New activity in jondurbin/airoboros-13b about 1 year ago

prompt format

#3 opened about 1 year ago by

underlines

New activity in BlackSamorez/llama-13b-hf-tensor-parallel about 1 year ago

Example of how to use

3

#1 opened about 1 year ago by

milyiyo

New activity in baseten/alpaca-30b about 1 year ago

Model can only run on 40G+ vram GPUS.

#1 opened about 1 year ago by

New activity in pszemraj/flan-t5-large-grammar-synthesis about 1 year ago

Strange output from: "I took my dauter to feed the pigeons today."

#8 opened about 1 year ago by