grok-1
grok-1 copied to clipboard
xai-org
Grok open release
Dear all, I could setup a DGX-1 (8x A100) for grok-1, with a lot of difficulties in the env. Still, all seems fine now, but at some point it just...
Adding comments to explain model configuration settings more clearly. Using an f-string for the error message logging to maintain consistency. Wrapping the model initialization and inference call in try-except blocks...
This refined version focuses on the advanced configurations such as the Transformer model setup with its large embedding size, the use of a Mixture of Experts (MoE) for increased model...
Hi thank you for opensourcing your model. However, I got the following error when I run run.py: `FileNotFoundError: [Errno 2] No such file or directory: 'D:\\dev\\shm\\tmpqye6hukj'` Note that I am...
dan@MacBook-Pro grok-1 % python3.11 run.py INFO:jax._src.xla_bridge:Unable to initialize backend 'cuda': INFO:jax._src.xla_bridge:Unable to initialize backend 'rocm': module 'jaxlib.xla_extension' has no attribute 'GpuAllocatorConfig' INFO:jax._src.xla_bridge:Unable to initialize backend 'tpu': INTERNAL: Failed to open...
To rapidly onboard new developers its best to provide a reproducible development environment. This PR achieves the following: - deterministic system dependencies using `nixpkgs` via `./flake.nix` - automatic repo setup...
Using more context managers, as well as dynamic temp dir creation, improve temp file handling, error handling, and logging
Instead of marking `value` for deletion by doing ```py del value # Unused. ``` ignore the variable in the function signature.
The key changes: - 1.) Validate checkpoint integrity by comparing hashes - 2.) Add rate limiting on inferences - 3.) Use authentication for any inference endpoints - 4.) Other general...
