atlasia/FineWeb2-Moroccan-Arabic-Predictions-model_multi_iso_v3.1_all_data_0.8fpr.bin
Updated
AtlasIA
community
AI & ML interests
Open-source data and models for Morocco.
Recent Activity
View all activity
atlasia's activity
BounharAbdelazizΒ
updated
a
dataset
about 2 hours ago
nouamanetaziΒ
updated
a
dataset
about 6 hours ago
choukraniΒ
updated
a
dataset
about 24 hours ago
imomayizΒ
updated
a
dataset
1 day ago
atlasia-maΒ
updated
a
dataset
1 day ago
nouamanetaziΒ
updated
a
collection
1 day ago
imomayizΒ
updated
2
collections
2 days ago
imomayizΒ
updated
a
dataset
2 days ago
abdeljalilELmajjodiΒ
updated
a
Space
3 days ago
BounharAbdelazizΒ
updated
a
Space
4 days ago
alielfilali01Β
posted
an
update
4 days ago
Post
1703
3C3H AraGen Leaderboard welcomes today
deepseek-ai/DeepSeek-V3 and 12 other models (including the late gpt-3.5 π) to the ranking of best LLMs in Arabic !
Observations:
- DeepSeek-v3 ranked 3rd and only Open model among the top 5 !
- A 14B open model ( Qwen/Qwen2.5-14B-Instruct) outperforms gpt-3.5-turbo-0125 (from last year). This shows how much we came in advancing and supporting Arabic presence within the LLM ecosystem !
- Contrary to what observed in likelihood-acc leaderboards (like OALL/Open-Arabic-LLM-Leaderboard) further finetuned models like maldv/Qwentile2.5-32B-Instruct actually decreased the performance compared to the original model Qwen/Qwen2.5-32B-Instruct.
It's worth to note that the decrease is statiscally insignificant which imply that at best, the out-domain finetuning do not really hurts the model original capabilities acquired during pretraining.
Previous work addressed this (finetuning VS pretraining) but more investigation in this regard is required (any PhDs here ? This could be your question ...)
Check out the latest rankings: inceptionai/AraGen-Leaderboard
Observations:
- DeepSeek-v3 ranked 3rd and only Open model among the top 5 !
- A 14B open model ( Qwen/Qwen2.5-14B-Instruct) outperforms gpt-3.5-turbo-0125 (from last year). This shows how much we came in advancing and supporting Arabic presence within the LLM ecosystem !
- Contrary to what observed in likelihood-acc leaderboards (like OALL/Open-Arabic-LLM-Leaderboard) further finetuned models like maldv/Qwentile2.5-32B-Instruct actually decreased the performance compared to the original model Qwen/Qwen2.5-32B-Instruct.
It's worth to note that the decrease is statiscally insignificant which imply that at best, the out-domain finetuning do not really hurts the model original capabilities acquired during pretraining.
Previous work addressed this (finetuning VS pretraining) but more investigation in this regard is required (any PhDs here ? This could be your question ...)
Check out the latest rankings: inceptionai/AraGen-Leaderboard
nouamanetaziΒ
updated
a
Space
7 days ago
alielfilali01Β
posted
an
update
12 days ago
Post
1805
~75% on the challenging GPQA with only 40M parameters π₯π₯³
GREAT ACHIEVEMENT ! Or is it ?
This new Work, "Data Laundering: Artificially Boosting Benchmark Results through Knowledge Distillation", take out the mystery about many models i personally suspected their results. Speacially on leaderboards other than the english one, Like the Open Arabic LLM Leaderbaord OALL/Open-Arabic-LLM-Leaderboard.
The authors of this work, first started by training a model on the GPQA data, which, unsurprisingly, led to the model achieving 100% performance.
Afterward, they trained what they referred to as a 'legitimate' model on legitimate data (MedMCQA). However, they introduced a distillation loss from the earlier, 'cheated' model.
What they discovered was fascinating: the knowledge of GPQA leaked through this distillation loss, even though the legitimate model was never explicitly trained on GPQA during this stage.
This raises important questions about the careful use of distillation in model training, especially when the training data is opaque. As they demonstrated, itβs apparently possible to (intentionally or unintentionally) leak test data through this method.
Find out more: Data Laundering: Artificially Boosting Benchmark Results through Knowledge Distillation (2412.15255)
GREAT ACHIEVEMENT ! Or is it ?
This new Work, "Data Laundering: Artificially Boosting Benchmark Results through Knowledge Distillation", take out the mystery about many models i personally suspected their results. Speacially on leaderboards other than the english one, Like the Open Arabic LLM Leaderbaord OALL/Open-Arabic-LLM-Leaderboard.
The authors of this work, first started by training a model on the GPQA data, which, unsurprisingly, led to the model achieving 100% performance.
Afterward, they trained what they referred to as a 'legitimate' model on legitimate data (MedMCQA). However, they introduced a distillation loss from the earlier, 'cheated' model.
What they discovered was fascinating: the knowledge of GPQA leaked through this distillation loss, even though the legitimate model was never explicitly trained on GPQA during this stage.
This raises important questions about the careful use of distillation in model training, especially when the training data is opaque. As they demonstrated, itβs apparently possible to (intentionally or unintentionally) leak test data through this method.
Find out more: Data Laundering: Artificially Boosting Benchmark Results through Knowledge Distillation (2412.15255)
- 1 reply
alielfilali01Β
posted
an
update
29 days ago
Post
3412
Unpopular opinion: Open Source takes courage to do !
Not everyone is brave enough to release what they have done (the way they've done it) to the wild to be judged !
It really requires a high level of "knowing wth are you doing" ! It's kind of a super power !
Cheers to the heroes here who see this!
Not everyone is brave enough to release what they have done (the way they've done it) to the wild to be judged !
It really requires a high level of "knowing wth are you doing" ! It's kind of a super power !
Cheers to the heroes here who see this!
-
- 3 replies
alielfilali01Β
posted
an
update
about 1 month ago
Post
1517
Apparently i forgot to put this here !
Well, this is a bit late but consider given our recent blog a read if you are interested in Evaluation.
You don't have to be into Arabic NLP in order to read it, the main contribution we are introducing is a new evaluation measure for NLG. We made the fisrt application of this measure on Arabic for now and we will be working with colleagues from the community to expand it to other languages.
Blog:
Rethinking LLM Evaluation with 3C3H: AraGen Benchmark and Leaderboard
https://huggingface.co/blog/leaderboard-3c3h-aragen
Space:
inceptionai/AraGen-Leaderboard
Give it a read and let me know your thoughts π€
Well, this is a bit late but consider given our recent blog a read if you are interested in Evaluation.
You don't have to be into Arabic NLP in order to read it, the main contribution we are introducing is a new evaluation measure for NLG. We made the fisrt application of this measure on Arabic for now and we will be working with colleagues from the community to expand it to other languages.
Blog:
Rethinking LLM Evaluation with 3C3H: AraGen Benchmark and Leaderboard
https://huggingface.co/blog/leaderboard-3c3h-aragen
Space:
inceptionai/AraGen-Leaderboard
Give it a read and let me know your thoughts π€
alielfilali01Β
posted
an
update
2 months ago
Post
2191
Unpopular opinion : o1-preview is more stupid than 4o and Qwen2.5-72B-Instruct in extremely underrated !
- 2 replies
alielfilali01Β
posted
an
update
3 months ago
Post
1706
I feel like this incredible resource hasn't gotten the attention it deserves in the community!
@clefourrier and generally the HuggingFace evaluation team put together a fantastic guidebook covering a lot about ππ©πππ¨ππ§ππ’π‘ from basics to advanced tips.
link : https://github.com/huggingface/evaluation-guidebook
I havenβt finished it yet, but i'am enjoying every piece of it so far. Huge thanks @clefourrier and the team for this invaluable resource !
@clefourrier and generally the HuggingFace evaluation team put together a fantastic guidebook covering a lot about ππ©πππ¨ππ§ππ’π‘ from basics to advanced tips.
link : https://github.com/huggingface/evaluation-guidebook
I havenβt finished it yet, but i'am enjoying every piece of it so far. Huge thanks @clefourrier and the team for this invaluable resource !
-
-
- 3 replies