Jun Young Baek's picture

1 2 75

Jun Young Baek

jupiterbjy

·

jupiterbjy

AI & ML interests

None yet

Recent Activity

replied to bartowski's post 3 days ago

Looks like Q4_0_N_M file types are going away Before you panic, there's a new "preferred" method which is online (I prefer the term on-the-fly) repacking, so if you download Q4_0 and your setup can benefit from repacking the weights into interleaved rows (what Q4_0_4_4 was doing), it will do that automatically and give you similar performance (minor losses I think due to using intrinsics instead of assembly, but intrinsics are more maintainable) You can see the reference PR here: https://github.com/ggerganov/llama.cpp/pull/10446 So if you update your llama.cpp past that point, you won't be able to run Q4_0_4_4 (unless they add backwards compatibility back), but Q4_0 should be the same speeds (though it may currently be bugged on some platforms) As such, I'll stop making those newer model formats soon, probably end of this week unless something changes, but you should be safe to download and Q4_0 quants and use those ! Also IQ4_NL supports repacking though not in as many shapes yet, but should get a respectable speed up on ARM chips, PR for that can be found here: https://github.com/ggerganov/llama.cpp/pull/10541 Remember, these are not meant for Apple silicon since those use the GPU and don't benefit from the repacking of weights

replied to bartowski's post 4 days ago

Looks like Q4_0_N_M file types are going away Before you panic, there's a new "preferred" method which is online (I prefer the term on-the-fly) repacking, so if you download Q4_0 and your setup can benefit from repacking the weights into interleaved rows (what Q4_0_4_4 was doing), it will do that automatically and give you similar performance (minor losses I think due to using intrinsics instead of assembly, but intrinsics are more maintainable) You can see the reference PR here: https://github.com/ggerganov/llama.cpp/pull/10446 So if you update your llama.cpp past that point, you won't be able to run Q4_0_4_4 (unless they add backwards compatibility back), but Q4_0 should be the same speeds (though it may currently be bugged on some platforms) As such, I'll stop making those newer model formats soon, probably end of this week unless something changes, but you should be safe to download and Q4_0 quants and use those ! Also IQ4_NL supports repacking though not in as many shapes yet, but should get a respectable speed up on ARM chips, PR for that can be found here: https://github.com/ggerganov/llama.cpp/pull/10541 Remember, these are not meant for Apple silicon since those use the GPU and don't benefit from the repacking of weights

reacted to bartowski's post with 👍 4 days ago

Looks like Q4_0_N_M file types are going away Before you panic, there's a new "preferred" method which is online (I prefer the term on-the-fly) repacking, so if you download Q4_0 and your setup can benefit from repacking the weights into interleaved rows (what Q4_0_4_4 was doing), it will do that automatically and give you similar performance (minor losses I think due to using intrinsics instead of assembly, but intrinsics are more maintainable) You can see the reference PR here: https://github.com/ggerganov/llama.cpp/pull/10446 So if you update your llama.cpp past that point, you won't be able to run Q4_0_4_4 (unless they add backwards compatibility back), but Q4_0 should be the same speeds (though it may currently be bugged on some platforms) As such, I'll stop making those newer model formats soon, probably end of this week unless something changes, but you should be safe to download and Q4_0 quants and use those ! Also IQ4_NL supports repacking though not in as many shapes yet, but should get a respectable speed up on ARM chips, PR for that can be found here: https://github.com/ggerganov/llama.cpp/pull/10541 Remember, these are not meant for Apple silicon since those use the GPU and don't benefit from the repacking of weights

View all activity

Organizations

None yet

jupiterbjy's activity

New activity in MaziyarPanahi/Llama-3-8B-Instruct-32k-v0.1-GGUF 6 months ago

Do you plan to provide a GGUF update with the BPE tokenizer fix?

#8 opened 8 months ago by