Björn Plüster's picture

Björn Plüster

bjoernp

·

AI & ML interests

None yet

Recent Activity

liked a dataset about 2 months ago

reyavir/PromptEvals

liked a dataset about 2 months ago

quotientai/HalluMix

liked a dataset 4 months ago

CohereLabs/include-base-44

View all activity

Organizations

bjoernp's activity

New activity in nvidia/Nemotron-4-340B-Base about 1 year ago

Hf safetensors version

#3 opened about 1 year ago by

New activity in LeoLM/leo-mistral-hessianai-7b-chat about 1 year ago

use_flash_attention_2=True

#9 opened about 1 year ago by

leo-mistral-hessianai-7b-chat for privateGPT

#8 opened about 1 year ago by

New activity in DiscoResearch/Llama3_DiscoLM_German_8b_v0.1_experimental about 1 year ago

Update tokenizer_config.json

#1 opened about 1 year ago by

New activity in LeoLM/leo-hessianai-13b-chat over 1 year ago

Problems with flash-attention2

#13 opened over 1 year ago by

New activity in DiscoResearch/mixtral-7b-8expert over 1 year ago

Loss function?

#10 opened over 1 year ago by

New activity in mistralai/Mixtral-8x7B-v0.1 over 1 year ago

No multi GPU inference support?

#4 opened over 1 year ago by

New activity in LeoLM/leo-hessianai-7b over 1 year ago

Llama2 vs Mistral

#2 opened over 1 year ago by

New activity in DiscoResearch/mixtral-7b-8expert over 1 year ago

Add languages

#8 opened over 1 year ago by

Missing module/classes: from transformers.cache_utils import Cache, DynamicCache

#7 opened over 1 year ago by

New activity in DiscoResearch/DiscoLM-mixtral-8x7b-v2 over 1 year ago

changed "tokenizer" typo to be the one we create.

#4 opened over 1 year ago by

Which transformers version is being used here?

#6 opened over 1 year ago by

Promptengineering

New activity in DiscoResearch/mixtral-7b-8expert over 1 year ago

Flash dependency (locks out non-NVIDIA GPUs)

#4 opened over 1 year ago by

Really appreciate the work put into this! I have noticed a change in the model output since first release.

#3 opened over 1 year ago by

New activity in DiscoResearch/DiscoLM-mixtral-8x7b-v2 over 1 year ago

Trying to quantize. Running into the issue below. Any suggestions?

#5 opened over 1 year ago by

small readme fix

#1 opened over 1 year ago by

New activity in DiscoResearch/mixtral-7b-8expert over 1 year ago

Update modeling_moe_mistral.py

#1 opened over 1 year ago by

New activity in LeoLM/leo-hessianai-70b-chat over 1 year ago

AWQ-Variante

#2 opened over 1 year ago by

Little Mistake :)

#1 opened over 1 year ago by

New activity in LeoLM/leo-hessianai-13b-chat over 1 year ago

Can you incorporate madlad400 training data ?

#11 opened over 1 year ago by