lhl's picture

lhl PRO

leonardlin

·

https://randomfoo.net/

lhl
lhl

AI & ML interests

None yet

Recent Activity

liked a model 1 day ago

featherless-ai/Qwerky-72B

posted an update 2 days ago

I'm excited to announce the official release of our Shisa V2 405B model: https://huggingface.co/shisa-ai/shisa-v2-llama3.1-405b It's the strongest model ever trained in Japan, and even goes toe-to-toe w/ GPT-4o and DeepSeek-V3 in JA MT-Bench. For all the details, be sure to check out post and overview report here: https://shisa.ai/posts/shisa-v2-405b/

updated a collection 2 days ago

View all activity

Organizations

leonardlin's activity

New activity in shisa-ai/ablation-191-finalsft2-shisa-v2-qwen2.5-7b about 1 month ago

Improve language tag

#1 opened about 1 month ago by

New activity in shisa-ai/shisa-v2-qwen2.5-7b about 1 month ago

Improve language tag

#1 opened about 1 month ago by

New activity in shisa-ai/ablation-194-finalsft2-shisa-v2-qwen2.5-32b about 1 month ago

Improve language tag

#1 opened about 1 month ago by

New activity in shisa-ai/shisa-v2-qwen2.5-32b about 1 month ago

Improve language tag

#2 opened about 1 month ago by

Improve language tag

#1 opened about 1 month ago by

New activity in mistralai/Mistral-Large-Instruct-2411 3 months ago

Broken tokenizer output on Japanese benchmarks?

#14 opened 3 months ago by

New activity in shisa-ai/shisa-v1-llama3-70b-gguf about 1 year ago

GGUF Issues

#1 opened about 1 year ago by

New activity in mlabonne/OmniBeagle-7B over 1 year ago

MT-Bench Scores

#1 opened over 1 year ago by

New activity in ShinojiResearch/Senku-70B-Full over 1 year ago

MT-Bench Results

#2 opened over 1 year ago by

New activity in mlabonne/NeuralBeagle14-7B over 1 year ago

Are there MT-Bench metrics produced for this model?

#4 opened over 1 year ago by

New activity in OrionStarAI/Orion-14B-Chat over 1 year ago

Add chat_template to tokenizer_config.json?

#1 opened over 1 year ago by

New activity in augmxnt/shisa-gamma-7b-v1 over 1 year ago

Does this tokenizer work?

#1 opened over 1 year ago by

New activity in tokyotech-llm/Swallow-7b-instruct-hf over 1 year ago

tokenizer.chat_template

#2 opened over 1 year ago by

New activity in augmxnt/shisa over 1 year ago

Apply for community grant: Personal project (gpu and storage)

#1 opened over 1 year ago by

New activity in augmxnt/shisa-7b-v1 over 1 year ago

Quants

#2 opened over 1 year ago by

llama.cpp doesn't work w/ Shisa, has a bug that affects certain BPE tokenizers like ours

#1 opened over 1 year ago by

New activity in augmxnt/shisa over 1 year ago

Tracking a few improvements TODO:

#2 opened over 1 year ago by

Apply for community grant: Personal project (gpu and storage)

#1 opened over 1 year ago by

Apply for community grant: Personal project (gpu and storage)

#1 opened over 1 year ago by