Ali El Filali's picture

Ali El Filali

alielfilali01

·

AI & ML interests

AI Psychometrician ? | NLP (mainly for Arabic) | Other interests include Reinforcement Learning and Cognitive sciences among others

Recent Activity

Reacted to vincentg64's post with 🧠 about 3 hours ago

There is no such thing as a Trained LLM https://mltblog.com/3CEJ9Pt What I mean here is that traditional LLMs are trained on tasks irrelevant to what they will do for the user. It’s like training a plane to efficiently operate on the runway, but not to fly. In short, it is almost impossible to train an LLM, and evaluating is just as challenging. Then, training is not even necessary. In this article, I dive on all these topics. ➡️ Training LLMs for the wrong tasks Since the beginnings with Bert, training an LLM typically consists of predicting the next tokens in a sentence, or removing some tokens and then have your algorithm fill the blanks. You optimize the underlying deep neural networks to perform these supervised learning tasks as well as possible. Typically, it involves growing the list of tokens in the training set to billions or trillions, increasing the cost and time to train. However, recently, there is a tendency to work with smaller datasets, by distilling the input sources and token lists. After all, out of one trillion tokens, 99% are noise and do not contribute to improving the results for the end-user; they may even contribute to hallucinations. Keep in mind that human beings have a vocabulary of about 30,000 keywords, and that the number of potential standardized prompts on a specialized corpus (and thus the number of potential answers) is less than a million. ➡️ Read the full articles at https://mltblog.com/3CEJ9Pt, also featuring issues with evaluation metrics and the benefits of untrained LLMs.

Reacted to malhajar's post with 🔥 3 days ago

🇫🇷 Lancement officiel de l'OpenLLM French Leaderboard : initiative open-source pour référencer l’évaluation des LLMs francophones Après beaucoup d’efforts et de sueurs avec Alexandre Lavallee, nous sommes ravis d’annoncer que le OpenLLMFrenchLeaderboard est en ligne sur Hugging Face (space url: https://huggingface.co/spaces/le-leadboard/OpenLLMFrenchLeaderboard) la toute première plateforme dédiée à l’évaluation des grands modèles de langage (LLM) en français. 🇫🇷✨ Ce projet de longue haleine est avant tout une œuvre de passion mais surtout une nécessité absolue. Il devient urgent et vital d'oeuvrer à plus de transparence dans ce domaine stratégique des LLM dits multilingues. La première pièce à l'édifice est donc la mise en place d'une évaluation systématique et systémique des modèles actuels et futurs. Votre modèle IA français est-il prêt à se démarquer ? Soumettez le dans notre espace, et voyez comment vous vous comparez par rapport aux autres modèles. ❓ Comment ça marche : Soumettez votre LLM français pour évaluation, et nous le testerons sur des benchmarks de référence spécifiquement adaptés pour la langue française — notre suite de benchmarks comprend : - BBH-fr : Raisonnement complexe - IFEval-fr : Suivi d'instructions - GPQA-fr : Connaissances avancées - MUSR-fr : Raisonnement narratif - MATH_LVL5-fr : Capacités mathématiques - MMMLU-fr : Compréhension multitâche Le processus est encore manuel, mais nous travaillons sur son automatisation, avec le soutien de la communauté Hugging Face. @clem , on se prépare pour une mise à niveau de l’espace ? 😏👀 Ce n'est pas qu'une question de chiffres—il s'agit de créer une IA qui reflète vraiment notre langue, notre culture et nos valeurs. OpenLLMFrenchLeaderboard est notre contribution personnelle pour façonner l'avenir des LLM en France.

View all activity

Articles

Introducing the Open Arabic LLM Leaderboard

Organizations

alielfilali01's activity

New activity in BounharAbdelaziz/AL-Atlas-Moroccan-Darija-Pretraining-Dataset 3 days ago

Damn that's cool !

#2 opened 3 days ago by

New activity in lhoestq/Common-Crawl-Pipeline-Creator about 1 month ago

Language filtering error

#1 opened about 2 months ago by

commented a paper about 2 months ago

Falcon Mamba: The First Competitive Attention-free 7B Language Model

Paper • 2410.05355 • Published Oct 7 • 29 •

New activity in NexaAIDev/Squid 3 months ago

"DolphinForCausalLM" OR "Qwen2ForCausalLM" ?

#3 opened 3 months ago by

New activity in CohereForAI/xP3x 3 months ago

What are the tasks ?

#10 opened 3 months ago by

New activity in upstage/open-ko-llm-leaderboard 4 months ago

Announcement: Launch of Open Ko-LLM Leaderboard Season 2

#83 opened 4 months ago by

New activity in OALL/Open-Arabic-LLM-Leaderboard 4 months ago

Some ‘finished’ models aren't visible in the leaderboard (result file is missing)

#5 opened 4 months ago by

Model eval request FAILED ... how do we know the root cause?

#7 opened 4 months ago by

Cancel running float32 model (kevinpro/Hydra-LLaMA3-8B-0513-preview)

#6 opened 4 months ago by

New activity in meta-llama/Llama-3.1-405B 4 months ago

405B or 410B ?

#8 opened 4 months ago by

New activity in HuggingFaceFW/fineweb-edu 5 months ago

Resources on DataTrove ?

#10 opened 5 months ago by

New activity in OALL/requests 5 months ago

Update cloudyu/Yi-34Bx2-MoE-60B-DPO_eval_request_False_bfloat16_Original.json

#12 opened 5 months ago by

New activity in OALL/Open-Arabic-LLM-Leaderboard 6 months ago

Looks like someone else submitted sambanovasystems/SambaLingo-Arabic-Base with wrong precision

#2 opened 6 months ago by

News 1 : About delay in Finished evaluation

#3 opened 6 months ago by

There is a delay in model evaluation

#4 opened 6 months ago by

New activity in alielfilali01/jupyter-lab 6 months ago

Job disconnected after 1000 steps

#1 opened 6 months ago by

New activity in OALL/requests 6 months ago

[bot] Conversion to Parquet

#1 opened 7 months ago by

parquet-converter

Create Qwen2-0.5B-Instruct_eval_request_False_bfloat16_Original.json

#8 opened 6 months ago by

Create Qwen2-1.5B-Instruct_eval_request_False_bfloat16_Original.json

#9 opened 6 months ago by

Create Qwen2-57B-A14B-Instruct_eval_request_False_bfloat16_Original.json

#10 opened 6 months ago by