grok-1
grok-1 copied to clipboard
xai-org
Grok open release
Hello I am an amateur AI behavioural researcher. I am currently working with Claude 3 I have a project I would like to export to grok because of the nature...
Best fix this quickly before you start having a flood of dependency issues raised. I created this ticket; https://github.com/xai-org/grok-1/issues/121 but it was closed by overzealous mods. It needs to be...
Hello, I use python3.10 to execute "python run.py", but it has an error: _pickle.UnpicklingError: invalid load key, '\x00'.
I've been playing around and noticed something interesting about how the model's memory gets used during inference. It looks like we're loading the model in FP16, but then, I saw...
confused about output_multiplier_scale, from the code they will multiply an output_multiplier_scale to logits before output. and that output_multiplier=0.5773502691896257, same as apply a default temperature with sqrt(3) but why? quite confused,...
Run on PC
Maybe stupid question, but how many RAM, VRAM and what processor need to run this :D
Are there plans to develop on the pytorch framework? Many LLM develop based on torch, like LLama, Qwen, etc. Some developers may not be familiar with JAX.
Can we download the original (supposedly bfloat16) weights for fine-tuning? The checkpoint is int8 quantized.
I as a hobbyist am not able to run a 300B model realistically, and fine tuning such a large model is likely even harder. I think Grok 0 weights should...
1. How to train, evaluate, and provide the train script 2. Example train datasets 3. Docs of grok's technical details.
