Michael Royzen's picture

13 81

Michael Royzen PRO

michaelroyzen

·

AI & ML interests

None yet

Recent Activity

liked a model 18 days ago

tencent/HunyuanVideo

liked a model 19 days ago

meta-llama/Llama-3.3-70B-Instruct

liked a Space about 1 month ago

black-forest-labs/FLUX.1-dev

View all activity

Organizations

michaelroyzen's activity

New activity in Phind/Phind-CodeLlama-34B-v2 over 1 year ago

Wizardcoder controversy

#2 opened over 1 year ago by

Issue/Bug replicating HumanEval result

#4 opened over 1 year ago by

Phind-CodeLlama-13b model?

#3 opened over 1 year ago by

I have some proof

#1 opened over 1 year ago by

New activity in Phind/Phind-CodeLlama-34B-Python-v1 over 1 year ago

This might be the best coding model (Better than wizardcoder_python_34b)

#5 opened over 1 year ago by

This is an absolute gem! Can't thank you enough.

#4 opened over 1 year ago by

New activity in Phind/Phind-CodeLlama-34B-v1 over 1 year ago

Python-only or multi-lingual?

#2 opened over 1 year ago by

New activity in Phind/Phind-CodeLlama-34B-Python-v1 over 1 year ago

True humaneval score or NewHope repeat?

#1 opened over 1 year ago by

New activity in Phind/Phind-CodeLlama-34B-v1 over 1 year ago

instruct-answer format?

#1 opened over 1 year ago by

New activity in bigcode/the-stack-dedup over 1 year ago

Bug: datasets.builder.DatasetGenerationError: An error occurred while generating the dataset

#30 opened over 1 year ago by

New activity in google/flan-t5-xxl almost 2 years ago

Flan-T5 tokenizer supports neither Chinese nor many code-related tokens despite being advertised as such

#33 opened almost 2 years ago by

New activity in google/flan-t5-xxl about 2 years ago

Only BF16 Work. FP16 and 8INT generate non-sense for me currently.

#9 opened about 2 years ago by

Activation function config inconsistent

#11 opened about 2 years ago by

Activation function config inconsistent

#11 opened about 2 years ago by

Activation function config inconsistent

#11 opened about 2 years ago by