Toolmaker. Software creator, optimizer and harmonizer. Guide and Book Author. Makes things work.
Current domains: LLM/Scalability/NLP/Machine Learning.
-
Stasosphere Online Inc. / HuggingFace Inc.
- Lantzville, BC, Canada
- https://stasosphere.com/machine-learning/
- @StasBekman
Block or Report
Block or report stas00
Report abuse
Contact GitHub support about this user’s behavior. Learn more about reporting abuse.
Report abusePinned
-
ipyexperiments Public
jupyter/ipython experiment containers for GPU and general RAM re-use
-
1,119 contributions in the last year
Less
More
Activity overview
Contributed to
huggingface/transformers,
microsoft/DeepSpeed,
stas00/toolbox
and 28 other
repositories
Contribution activity
June 2023
Created 9 commits in 3 repositories
Created a pull request in huggingface/transformers that received 3 comments
[tests] fix bitsandbytes import issue
this test has been failing when peft library was installed since it tries to access bitsandbytes.nn
$ pip install accelerate peft bitsandbytes==0.3…
+0
−0
•
3
comments
Reviewed 1 pull request in 1 repository
huggingface/accelerate
1 pull request
Created an issue in huggingface/accelerate that received 5 comments
LR resume is somewhat buggy
System Info During m4 training we had to hack the LR resume state file to adjust LR in the middle of the training. As I wasn't sure if I did the ri…
2 of 4 tasks
19
contributions
in private repositories
Jun 3 – Jun 15





