schmorp
schmorp
Seeing how much jpgtran is in use, good command line tools will go a long way to cement jpeg-xl for decades :) Especially things that are very hard with jpeg-xl...
Sorry to barge in, but I would urge the jpeg xl maintainers to point out these limitations clearly, because currently you can read everyhwere that jpeg-xl can losslessly transcode to...
This happens for hyperscan 5.4.0 and 5.3.0 regardless of fat binary support or not. Hyperscan 5.2.1 is the last release that builds for me.
I hope you didn't ask me for the wording - but for what its worth that wording looks just fine to me :)
Actually, by now, only 5 of the 8 jobs failed, so it clearly depends on the model and is not universal.
Just noticed this during another imatrix calculation: [20]17.0770,[21]17.3345,[22]17.2957,[23]17.5153,[24]17.4855,[25]17.4871,[26]17.4550,[27]17.4515,[28]17.3787,[29]17.2030,[30]17.1163,[31]16.9776,[32]16.9705,[33]16.8447,[34]16.7977,[35]16.6126,[36]16.4577,[37]16.3338,[38]16.1736,[39]16.2237, save_imatrix: stored collected data after 40 chunks in miqu-1-120b.imatrix~ [40]16.3695,[41]16.2985,[42]16.2147,[43]16.1276,[44]16.1080,[45]nan,[46]nan,[47]nan,[48]nan,[49]nan,[50]nan,[51]nan,[52]nan,[53]nan,[54]nan,[55]nan,[56]nan,[57]nan Could be incidental, but I don't think I have noticed "nan" values...
miqu-1-120b failed the same way. I'm quantizing models for quite a while now, many miqu variants made the same way. It's very unlikely that I just happened ot pick invalid...
save_imatrix: stored collected data after 40 chunks in miqu-1-120b.imatrix~ [40]16.3695,[41]16.2985,[42]16.2147,[43]16.1276,[44]16.1080,[45]15.9874,[46]16.0110,[47]15.9119, nan's do not appear with b2355.
With the Q8 quant from here: https://huggingface.co/mradermacher/BagelMix-8x7B-GGUF And this invocation: imatrix -ofreq 20 -t 4 -ngl 0 -mg 0 -m ... -o ... -f ... I reliably get nan after...
Actually, I wrote this when the test with group_10_merged.txt was at the end, but it did give a nan result near the end with (https://github.com/ggerganov/llama.cpp/files/14143306/group_10_merged.txt btw.)