Running
45
š„
Attention Visualization
Vision Transformer Attention Visualization
Akarshan Biswas
qnixsynapse
AI & ML interests
NLP, models, quantization
Recent Activity
new activity
4 days ago
google/gemma-2-9b-it:Tool calling support in Gemma 2
liked
a Space
4 days ago
webml-community/attention-visualization
reacted
to
suayptalha's
post
with š
7 days ago
š Introducing š
š¢š«š¬š šš®š š š¢š§š š
ššš šš§ššš š«ššš¢šØš§ šØš š¦š¢š§ššš ššØššš„š¬ from the paper ššš«š šššš¬ šš„š„ šš šššššš?
š„ I have integrated š§šš±š-š šš§šš«ššš¢šØš§ šššš¬, specifically minGRU, which offer faster performance compared to Transformer architectures, into HuggingFace. This allows users to leverage the lighter and more efficient minGRU models with the "šš«šš§š¬ššØš«š¦šš«š¬" š„š¢šš«šš«š² for both usage and training.
š» I integrated two main tasks: šš¢š§šššš
šØš«šššŖš®šš§šššš„šš¬š¬š¢šš¢šššš¢šØš§ and šš¢š§šššš
šØš«ššš®š¬šš„šš.
šš¢š§šššš
šØš«šššŖš®šš§šššš„šš¬š¬š¢šš¢šššš¢šØš§:
You can use this class for šššŖš®šš§šš šš„šš¬š¬š¢šš¢šššš¢šØš§ tasks. I also trained a Sentiment Analysis model with stanfordnlp/imdb dataset.
šš¢š§šššš
šØš«ššš®š¬šš„šš:
You can use this class for ššš®š¬šš„ ššš§š š®šš š ššØššš„ tasks such as GPT, Llama. I also trained an example model with roneneldan/TinyStories dataset. You can fine-tune and use it!
š šš¢š§š¤š¬:
Models: https://huggingface.co/collections/suayptalha/mingru-676fe8d90760d01b7955d7ab
GitHub: https://github.com/suayptalha/minGRU-hf
LinkedIn Post: https://www.linkedin.com/posts/suayp-talha-kocabay_mingru-a-suayptalha-collection-activity-7278755484172439552-wNY1
š° šš«ššš¢šš¬:
Paper Link: https://arxiv.org/abs/2410.01201
I am thankful to Leo Feng, Frederick Tung, Mohamed Osama Ahmed, Yoshua Bengio and Hossein Hajimirsadeghi for their papers.
Organizations
None yet
qnixsynapse's activity
Tool calling support in Gemma 2
2
#50 opened 27 days ago
by
qnixsynapse
liked
a
Space
4 days ago
reacted to
suayptalha's
post with š
7 days ago
Post
1785
š Introducing š
š¢š«š¬š šš®š š š¢š§š š
ššš šš§ššš š«ššš¢šØš§ šØš š¦š¢š§ššš ššØššš„š¬ from the paper ššš«š šššš¬ šš„š„ šš šššššš?
š„ I have integrated š§šš±š-š šš§šš«ššš¢šØš§ šššš¬, specifically minGRU, which offer faster performance compared to Transformer architectures, into HuggingFace. This allows users to leverage the lighter and more efficient minGRU models with the "šš«šš§š¬ššØš«š¦šš«š¬" š„š¢šš«šš«š² for both usage and training.
š» I integrated two main tasks: šš¢š§šššš šØš«šššŖš®šš§šššš„šš¬š¬š¢šš¢šššš¢šØš§ and šš¢š§šššš šØš«ššš®š¬šš„šš.
šš¢š§šššš šØš«šššŖš®šš§šššš„šš¬š¬š¢šš¢šššš¢šØš§:
You can use this class for šššŖš®šš§šš šš„šš¬š¬š¢šš¢šššš¢šØš§ tasks. I also trained a Sentiment Analysis model with stanfordnlp/imdb dataset.
šš¢š§šššš šØš«ššš®š¬šš„šš:
You can use this class for ššš®š¬šš„ ššš§š š®šš š ššØššš„ tasks such as GPT, Llama. I also trained an example model with roneneldan/TinyStories dataset. You can fine-tune and use it!
š šš¢š§š¤š¬:
Models: suayptalha/mingru-676fe8d90760d01b7955d7ab
GitHub: https://github.com/suayptalha/minGRU-hf
LinkedIn Post: https://www.linkedin.com/posts/suayp-talha-kocabay_mingru-a-suayptalha-collection-activity-7278755484172439552-wNY1
š° šš«ššš¢šš¬:
Paper Link: https://arxiv.org/abs/2410.01201
I am thankful to Leo Feng, Frederick Tung, Mohamed Osama Ahmed, Yoshua Bengio and Hossein Hajimirsadeghi for their papers.
š„ I have integrated š§šš±š-š šš§šš«ššš¢šØš§ šššš¬, specifically minGRU, which offer faster performance compared to Transformer architectures, into HuggingFace. This allows users to leverage the lighter and more efficient minGRU models with the "šš«šš§š¬ššØš«š¦šš«š¬" š„š¢šš«šš«š² for both usage and training.
š» I integrated two main tasks: šš¢š§šššš šØš«šššŖš®šš§šššš„šš¬š¬š¢šš¢šššš¢šØš§ and šš¢š§šššš šØš«ššš®š¬šš„šš.
šš¢š§šššš šØš«šššŖš®šš§šššš„šš¬š¬š¢šš¢šššš¢šØš§:
You can use this class for šššŖš®šš§šš šš„šš¬š¬š¢šš¢šššš¢šØš§ tasks. I also trained a Sentiment Analysis model with stanfordnlp/imdb dataset.
šš¢š§šššš šØš«ššš®š¬šš„šš:
You can use this class for ššš®š¬šš„ ššš§š š®šš š ššØššš„ tasks such as GPT, Llama. I also trained an example model with roneneldan/TinyStories dataset. You can fine-tune and use it!
š šš¢š§š¤š¬:
Models: suayptalha/mingru-676fe8d90760d01b7955d7ab
GitHub: https://github.com/suayptalha/minGRU-hf
LinkedIn Post: https://www.linkedin.com/posts/suayp-talha-kocabay_mingru-a-suayptalha-collection-activity-7278755484172439552-wNY1
š° šš«ššš¢šš¬:
Paper Link: https://arxiv.org/abs/2410.01201
I am thankful to Leo Feng, Frederick Tung, Mohamed Osama Ahmed, Yoshua Bengio and Hossein Hajimirsadeghi for their papers.
liked
a
model
29 days ago
upvoted
a
paper
about 1 month ago
liked
a
model
about 1 month ago
liked
a
Space
about 1 month ago
liked
a
model
2 months ago
liked
3
models
3 months ago
upvoted
a
paper
4 months ago
updated
a
model
4 months ago
Is this really an Instruct model?
#1 opened 4 months ago
by
qnixsynapse
liked
2
models
4 months ago
liked
a
dataset
4 months ago
liked
a
model
4 months ago
Depends upon the GPU hardware tbh. Not everywhere you can get 90 tokens/sec. :)
upvoted
a
paper
5 months ago