drunkinlove

@king-menin Perplexity at step 16k is around 160. I found `train_micro_batch_size_per_gpu=4` in the deepspeed config, is it supposed to equal the `batch-size` arg for pretrain_gpt3.py? Also, isn't overfitting supposed to...

Fixed that by installing the necessary dependencies through requirements.txt. This is the error I get now: ``` user@cloudshell:~/conversational-datasets (reddit-data-288210)$ python reddit/create_data.py \ > --output_dir ${DATADIR?} \ > --reddit_table ${PROJECT?}:${DATASET?}.${TABLE?} \...

I managed to make it work by using JSON and commenting out TensorFlow in both requirements.txt and create_data.py. Tried all TF versions from 1.11 to 1.15 otherwise to no success...

I have the same problem, unfortunately. M2, Ventura, freac 1.17, adding WavPack files

With the LinuxArm version, I get the following: `-bash: ./TwitchDownloaderCLI: cannot execute binary file: Exec format error` (which is weird, I thought it should run fine in compatibility mode?) Here's...

I tried using the x64 version of the app on Ubuntu with Apple virtualization framework (instead of qemu) and actually got a stack trace: ``` TwitchDownloaderCLI 1.53.4 Copyright (c) lay295...

@superbonaci hi! I'm on M2. hyfetch says the same (and I got a nice flag): ``` OS: Debian GNU/Linux 11 (bullseye) aarch64 CPU: 0x00 Architecture: aarch64 (4) ``` I installed...

@superbonaci I'm on Sonoma, yes, and I'm using the binary you linked. My Debian image is the ARM64 version from the official website. I'll try the one hosted by UTM...

@superbonaci I tried this and it worked! 🤔 This is the first app I've had problems with on the official Debian image; ffmpeg, python, mp4box and others ran fine. Although...

The armel/armhf are for 32-bit architectures: > Debian fully supports three ports to different flavours of little-endian ARM hardware: > > The [ARM EABI](https://wiki.debian.org/ArmEabiPort) (armel) port targets a range of...