Hugging Face's logo
Sanghoon Kim's picture
19 5

Sanghoon Kim

Limerobot
21world's profile picture BearSean's profile picture Yescia's profile picture
·
https://www.kaggle.com/limerobot

AI & ML interests

None yet

Organizations

upstage's profile picture illgamhoduck's profile picture UpstageShareSpace's profile picture Open Ko-LLM Leaderboard's profile picture TokenizerShareSpace's profile picture UpstageShareFireworks's profile picture UpstageShareTogether's profile picture KTShareSpace's profile picture UpstageSharePredibase's profile picture DLAIShareSpace's profile picture UpstageShareOpenPipe's profile picture UpstageShareJTS's profile picture

Limerobot's activity

New activity in upstage/open-ko-llm-leaderboard 10 months ago

GPT 평가 결과

2
#43 opened 10 months ago by maywell
New activity in upstage/SOLAR-10.7B-Instruct-v1.0 11 months ago

Add base model
#11 opened 11 months ago by osanseviero

It is just rambling

2
#10 opened 11 months ago by Mikolai007

This model ranked 1st place in open llm leader board, However this model has lower performance in supervised fine tuning.

2
#8 opened 11 months ago by andreaKIM

Use cache true

1
#6 opened 11 months ago by ParanjayChawla
New activity in upstage/open-ko-llm-leaderboard 11 months ago

Can I get the error log for my failed model?

3
#27 opened 12 months ago by heavytail

Are all metics in the table accuracy?

2
#28 opened 12 months ago by zhiminy
New activity in upstage/open-ko-llm-leaderboard about 1 year ago

Ko Common Gen V2 평가 방법 업데이트 안내
#12 opened about 1 year ago by Limerobot

Delay issue

1
#10 opened about 1 year ago by jyoung105

Update on the Evaluation Method for Ko Common Gen V2

2
#11 opened about 1 year ago by Limerobot

Model has failed.

2
#9 opened about 1 year ago by amphora
New activity in upstage/llama-30b-instruct-2048 over 1 year ago

Which one is better, finetuning with a sequence length of 1024 or 2048 in LLaMA1 and LLaMA2? And what are the reasons behind the choice?

2
#7 opened over 1 year ago by danielpark
New activity in upstage/Llama-2-70b-instruct over 1 year ago

Context window of this model?

1
#4 opened over 1 year ago by yiouyou
New activity in upstage/llama-30b-instruct-2048 over 1 year ago

Can it run faster than 2 tokens/second on one A100?

1
#6 opened over 1 year ago by aibarito-ua
New activity in open-llm-leaderboard/open_llm_leaderboard over 1 year ago

Our submitted model is still not visible on the leaderboard

10
#129 opened over 1 year ago by Limerobot
New activity in upstage/llama-65b-instruct over 1 year ago

tokenizer.model is missing

5
#4 opened over 1 year ago by YuxinXiao
New activity in upstage/Llama-2-70b-instruct over 1 year ago

Does this finetune let you use 4096 context?

1
#2 opened over 1 year ago by Panchovix
New activity in upstage/llama-30b-instruct over 1 year ago

What is this?

7
#2 opened over 1 year ago by TheBloke
New activity in upstage/llama-30b-instruct-2048 over 1 year ago

what is the prompt init to use

1
#4 opened over 1 year ago by pabloce

what is the prompt init to use

1
#4 opened over 1 year ago by pabloce