7 2

Iftach Arbel

iarbel

https://www.tenai.io/

iarbel84

AI & ML interests

Optimization Algorithms, NLP, LLMs, CV

Recent Activity

New activity about 1 month ago

cimphony-ai-admin/Cimphony-Mistral-Law-7B:Fail to run on vLLM

View all activity

Organizations

iarbel's activity

New activity in cimphony-ai-admin/Cimphony-Mistral-Law-7B about 1 month ago

Fail to run on vLLM

#1 opened about 2 months ago by

yaronr

updated a Space 3 months ago

Sleeping

🚀

Amazon Feature Bullets Demo

New activity in microsoft/Phi-3-mini-4k-instruct 4 months ago

Is it required to add a BOS token?

#92 opened 4 months ago by

iarbel

updated a Space 6 months ago

Sleeping

🕊️

Didi Annotation Space V2

replied to DmitryRyumin's post 6 months ago

The evaluation benchmarks use zero-shot, where usually few-shot is used. This raises the question whether the few-shot results weren't as good, compared to similar-sized models.

Hopefully the authors will release few-shot results, aligned with common practice (e.g. HF Open LLM Leaderboard)

commented a paper 6 months ago

You Only Cache Once: Decoder-Decoder Architectures for Language Models

Paper • 2405.05254 • Published May 8 • 9 •

updated 2 models 7 months ago

cimphony-ai-admin/Cimphony-Mistral-Law-7B

Text Generation • Updated May 14 • 5 • 2

iarbel/law-chat-base

Text Generation • Updated Apr 25 • 10

updated a Space 8 months ago

Sleeping

🕊️

Didi Annotation Space

updated 2 datasets 8 months ago

iarbel/unfair_tos_fewshot_eval

Viewer • Updated Mar 28 • 1.61k • 38

iarbel/legal_eval

Viewer • Updated Mar 28 • 14.6k • 115

New activity in Equall/Saul-7B-Instruct-v1 8 months ago

How was the evaluation executed?

#3 opened 8 months ago by

iarbel