Filter
Exclude
Time range
-
Near
@JAtanackov: Geomagnetic Storms r artificially induced to unrest earthquakes, earth dwellers are aware of that. Tho I'm not sure if you are aware that @Google can predict the tremors and divert accordingly. #Logfiles exist. 🌐Apache (lnkd.in/gjM-z_z5) | @UNGeneva | @NOAA
Replying to @JAtanackov
Geomagnetic activity rapidly reached G5 (extreme) storm level, DST min going down to -301 nT. This extreme storm became known as the Bastille day event (of 2000). One of three G5 storms of SC23.
23
nein du denkst viel zu komplex für enterprise IT. zu 99% klassisches overfetching im SQL, fehlender index oder so komplette brainrot-ideen wie logfiles in die produktionsdb schreiben.
1
1
17
Thank you for this result! Here's one initial correction: In this post, Rohan states that I made an attempt to implement Shampoo, and labels my Shampoo training run `Buggy shampoo impl`. But this is incorrect: I did not implement my own version of Shampoo. Instead -- as can be seen in the reproducible log I provided in the original post -- I used, out-of-the-box, the official DistributedShampoo implementation provided by Facebook research. This is the most commonly-used Shampoo implementation that I could find on the internet. The extent of my use of this implementation can be seen in the few lines of code below. If there are indeed bugs in this implementation, I can safely say that they were not created by me. The remaining mystery, for anyone interested, might naturally become something like the following - How did Rohan today achieve a significantly better result compared to the official 2022-era DistributedShampoo implementation? I am grateful for the comments he has already made regarding the deltas between his version and the 2022 one, and I am looking forward to things becoming fully precise/detailed soon once he releases the reproducible logfiles generated by his runs.
Here you go, sir. Muon is a good optimizer. I think Keller's attempt at implementing it is great -- this primarily helped me look at why his effort produced much worse looking curve than what I could get. This ended up being a nerdsnipe into hyper parameters, grafting, and eigh calls in the distributed shampoo package from Meta in my car ride home. The main delta's are below:
9
5
264
55,492
Replying to @_arohan_
Thanks for this result. Here is my initial comment while awaiting the reproducible logfiles of these runs

Thank you for this result; I appreciate it. Here's one initial correction: In his post, Rohan states that I made an attempt to implement Shampoo, and labels my Shampoo training run `Buggy shampoo impl`. But this is incorrect: I did not implement my own version of Shampoo. Instead -- as can be seen in the reproducible log I provided in the original post -- I used, out-of-the-box, the official DistributedShampoo implementation provided by Facebook research - for which it so happens that @_arohan_ has contributing credits. If there are indeed bugs in this official implementation, I can safely say that they were not created by me. The remaining mystery, for anyone interested, might naturally become something like the following - How did Rohan today achieve a significantly better result compared to the official 2022-era DistributedShampoo implementation? I am grateful for the comments he has already made regarding the deltas between his version and the 2022 one, and I am looking forward to things becoming fully precise/detailed soon once he releases the reproducible logfiles generated by his runs.
12
2,547
Thank you for this result; I appreciate it. Here's one initial correction: In his post, Rohan states that I made an attempt to implement Shampoo, and labels my Shampoo training run `Buggy shampoo impl`. But this is incorrect: I did not implement my own version of Shampoo. Instead -- as can be seen in the reproducible log I provided in the original post -- I used, out-of-the-box, the official DistributedShampoo implementation provided by Facebook research - for which it so happens that @_arohan_ has contributing credits. If there are indeed bugs in this official implementation, I can safely say that they were not created by me. The remaining mystery, for anyone interested, might naturally become something like the following - How did Rohan today achieve a significantly better result compared to the official 2022-era DistributedShampoo implementation? I am grateful for the comments he has already made regarding the deltas between his version and the 2022 one, and I am looking forward to things becoming fully precise/detailed soon once he releases the reproducible logfiles generated by his runs.
Here you go, sir. Muon is a good optimizer. I think Keller's attempt at implementing it is great -- this primarily helped me look at why his effort produced much worse looking curve than what I could get. This ended up being a nerdsnipe into hyper parameters, grafting, and eigh calls in the distributed shampoo package from Meta in my car ride home. The main delta's are below:
1
9
3,058
Replying to @patipvlr
srttrail.txt Windows 10 Fix" | How to Fix C /Windows/System32/LogFiles/srt/SrtTrail.txt joga isso no yt
82
竹下⚖ retweeted
Replying to @hasaka455 @Astrvhh
画像の状態は Windowsの起動ファイルが壊れている状態です。 画面の表示内容: 「スタートアップ修復でPCを修復できませんでした」 ログ:C:\Windows\System32\Logfiles\Srt\SrtTrail.txt つまり Windowsが正常に起動できないレベルでOSが破損しています。 原因として多いもの OSのシステムファイル破損 SSD / HDDのエラー 無理な最適化ツール・設定変更 Windows更新の失敗 電源強制終了の繰り返し 現在の状態 起動 → 10秒でフリーズ 起動時間が長くなった 自動修復失敗 これは Windowsが壊れている典型症状です。 対処方法(確実に直る方法) ① 画面の 「詳細オプション」 を押す ② トラブルシューティング ③ このPCを初期状態に戻す ④ 個人ファイルを保持する これで Windowsを再インストールします。 もしそれでもダメなら SSDが壊れている可能性が高いです。 その場合 SSD交換 Windows再インストール が必要になります。 結論 PC自体が壊れたわけではなく、Windows(OS)が破損している状態です。
2
3
37
38,892
Ich hab mich doch entschlossen die Intel Arc Pro B70 32GB zurück zu geben und für paar eur mehr die AMD R9700 zu nehmen. An vielen Stellen ist die Intel abgestürzt und mit viel Context extrem langsam geworden. Die R9700 ist teilweise im Prompt Processing und Token Generation 10 mal schneller und vor allem scheint sie extrem stabil zu sein. Die speziellen HIP (ROCm) Treiber scheinen nicht schneller zu sein, als die Standard Vulcan Treiber, so dass ich sogar noch Hoffnung habe, dass da in Zukunft was geht. Ich hab laut Logfiles letzten Monat ca 150€ als Sonnet 4.6 Äquivalent gespart, so dass der Invest recht schnell sich zurück zahlt. Vielleicht werde ich unvernünftig und kaufe noch eine zweite... Standard Sachen gehen mit Qwen3.6 35B mit Q5 teilweise schneller und besser als Sonnet 4.6. Schade, die Intel wäre echt cool gewesen, aber die Treiber / Software sehr schlecht.
10
1
26
2,917
Immer so schlecht recherchiert. Die Bundesbehörden sind sehr fahrlässig im Umgang mit den Daten der Bürger. So muss man zum Beispiel um die Datenschutz Erklärung auf den Bundes Seiten zu lesen, zuerst einmal die Suchfunktion bemühen. Hat man diese dann gefunden, erfährt man dort das einige Internetseiten des Bundes die Suchfunktion von Google verwenden, welche dies sind erfährt man dort jedoch nicht. Also eine klare Verletzung des Datenschutz. Mit jedem Formular das man auf den Seiten versendet gibt man Daten weiter. Dazu gibt es zwei Methoden nämlich GET oder POST. Bei GET sieht der Nutzer die Parameter welche weiter gegeben werden in der URL. Bei POST Versand sieht dies der Nutzer jedoch nicht, weil diese im HTTP-Header weitergegeben werden. Im HTTP Header werden jedoch auch Daten weitergegeben, welche vom Webserver dazu verwendet werden, um Logfiles zu erstellen. Doch nun kommt das schlimmste Szenario welches man sich datenschutzrechtlich vorstellen kann. Der gesamte Datenverkehr des Bundes, läuft über die Domainnameserver des Online-Händlers Amazon.
KI wird auch in unserem öffentlichen Raum immer wichtiger. Das gibt Big Tech neue Macht – und wirft Fragen auf. srf.ch/news/dialog/smart-cit…
2
5
18
533
May 18
Danke an die KI, das sie mir die Logfiles immer richtig auswertet und die Fehler in der Konfiguration findet 🤓
1
3
115
パソコン詳しい有識者教えてくれパソコン開けなくなっちまった エラーコードを置いておいらは寝ます X\windows\system32\logfiles\Srt\SrtTrail.txt kmode exception not handled
突然のブルスクに動揺が隠せません
1
2
557
Lost in the logfiles of my own making Wondering if I'm just patching up dreams
1
2
Replying to @michelportier80
Ik gebruik een DPN een hardware vpn. Ze kunnen vpn stoppen daar deze servers her en der staan. Met dpn zijn de gebruikers de server en logfiles blijft thuis op je apparaat.kan 10% korting geven.
1
3
424
Wer den Blick nach innen richtet, sieht nicht Schwäche, sondern die Logfiles der eigenen Realität 😂
1
2
21
@CoinbaseSupport I am receiving lots of meaningless ticker messages over the advanced API ticker channel since nearly two weeks. Tried to reach out to you without success. Posting it here. Although the application can handle meaningless tickers, there seems to be an issue for clients located in germany receiving meaningless tickers. Would be cool to silence the logfiles and SNMP traps. coinbase: No 'product_id' string item: {"type":"ticker","product_id":"","price":"","volume_24_h":"","low_24_h":"","high_24_h":"","low_52_w":"","high_52_w":"","price_percent_chg_24_h":"","best_bid":"94","best_ask":"96.492","best_bid_quantity":"1","best_ask_quantity":"25"}
11
4
61
Snap ik, maar dat wordt altijd vermeld in de voorwaarden, althans zijn ze dat verplicht. ODIDO heeft dit wat betreft de data vanuit de logfiles van hun modems destijds nagelaten te vermelden. Overigens is, naar eigen zeggen van ODIDO, dit nu aangepast en gebeurt dat niet meer.
3
92
Replying to @shiinozu
Also kann empfehlen wenn du mal akut nicht weißt, was es ist, deine Logfiles bei OBS auf der Seite hochzuladen. Der spuckt oft direkt aus, was das Problem ist. Aber auch leider nur manchmal. obsproject.com/tools/analyze…

2
13
637
I asked Claude Code (sonnet, medium) to add the results in some new logfiles to a .csv database, but instead I got 90k tokens of it verifying each calculation already in the dataset...
1
6
580