alpha 's picture

alpha

alphaprime90

·

ab0x

AI & ML interests

None yet

Organizations

None yet

alphaprime90's activity

New activity in TIGER-Lab/MMLU-Pro 4 months ago

Could you share the test result for each question?

#6 opened 4 months ago by

New activity in HeshamHaroon/Arabic-llama3 5 months ago

Could you talk about the changes you made to the model?

#2 opened 5 months ago by

New activity in MaziyarPanahi/Meta-Llama-3-8B-Instruct-GGUF 5 months ago

OK llama 3 8b model is INSANE. Is almost as good as wizard 2 8x22b!

#5 opened 5 months ago by

New activity in meta-llama/Meta-Llama-3-8B-Instruct 5 months ago

Non-English language capabilities

#2 opened 5 months ago by

New activity in cmp-nct/llava-1.6-gguf 7 months ago

mmproj files for smaller models mainly deepseek 1.3b and others.

#8 opened 7 months ago by

New activity in olafgeibig/phi-2-OpenHermes-2.5-GGUF 7 months ago

How does it compare to phi-2 or other small models?

#1 opened 7 months ago by

New activity in abacusai/TheProfessor-155b-gguf 7 months ago

To the few who can run it, Post speeds, setup, results quality.

#1 opened 7 months ago by

New activity in TheBloke/Wizard-Vicuna-13B-Uncensored-GGML over 1 year ago

New quant 8bit method, how is it performing on your CPU? (share your token/s, CPU model and -- thread)

#2 opened over 1 year ago by

New activity in CarperAI/stable-vicuna-13b-delta over 1 year ago

Can someone share ggml 4bit version?

#2 opened over 1 year ago by

New activity in togethercomputer/RedPajama-Data-1T over 1 year ago

What's the size of the data?

#5 opened over 1 year ago by

New activity in eachadea/ggml-vicuna-7b-1.1 over 1 year ago

Nice work. Very conversational, great formatting. Still hallucinates.

#1 opened over 1 year ago by

New activity in Sosaka/Alpaca-native-4bit-ggml over 1 year ago

Fine-tuning the model

#1 opened over 1 year ago by