goichi harada(webbigdata)

Hi. @hiromu166. I saw this project on Twitter. Can I do the chapter7 Main NLP tasks?

hi. I have stamped the discord channel. By the way, I submitted a pull request and it failed with the following error. ` files need to be formatted, run make...

for example, If you are installing diffusers from a clone of ``CompVis/stable-diffusion``, you will get this error. The file that exists on github is newer than the file that can...

Uninstall flash-attention once ``` pip uninstall flash-attn ``` Install flash-attention from source ``` git clone https://github.com/Dao-AILab/flash-attention cd flash-attention python setup.py install ``` works for me. https://github.com/Dao-AILab/flash-attention/issues/931

Hi, Thank you for your reply. The base model is https://huggingface.co/google/gemma-7b That is, it is a PT model. As far as I understand, if the prompt template is original, the...

Hi @KumarGitesh2024  ``` [ Reading prompt ] Entering GenerateGemma Entering GenerateImpl Entering RangeChecks Entering Prefill Entering Attention Entering FFW Entering Attention Entering FFW Entering Attention Entering FFW Entering Attention...

I found that not only EOT but also Japanese nor English was being output. ### google original 2b-it-sfp.sbs looks good. ``` ./build/gemma --tokenizer util/tokenizer.spm --compressed_weights util/2b-it-sfp.sbs --model 2b-it ``` log...

Hello. Unfortunately, there was no particular difference with the 16-bit version. Could you please confirm that I'm doing it right just to be sure? ``` cd build/ rm -rf *...

I recompiled it but the situation did not change. When I output the token ID in auto stream_token in run.cc, all the token IDs were 0. 133 } else {...

Thank you. I'm not in a hurry, so if you find out the cause, please let me know. If I can get gemma.cpp to work, it will open the door...