AMajor/grok-1
1
0
Fork 0
mirror of https://github.com/xai-org/grok-1.git synced 2024-11-23 03:59:53 +03:00
Code Issues Packages Projects Releases Wiki Activity

Compare commits

base: AMajor:e50578b5f50e4c10c6e7cff31af1ef2bedb3beb8
Branches Tags
AMajor:main
AMajor:download-instruction
..
compare: AMajor:d6d9447e2d3c9bd81da571dc8681ea60009e4c03
Branches Tags
AMajor:main
AMajor:download-instruction

5 Commits
e50578b5f5 ... d6d9447e2d

Author SHA1 Message Date
Szymon Tworkowski
d6d9447e2d
Update huggingface link 2024-03-18 11:40:01 -07:00
Lve Lvee
7207216386
Create .gitignore for checkpoints (#149) 
ignore the checkpoints files
 2024-03-18 11:01:17 -07:00
Seth Junot
310e19eee2
Corrected checkpoint dir name, download section link 2024-03-18 09:39:02 -07:00
Gareth Paul Jones (GPJ)
1ff4435d25
Update README with Model Specifications (#27) 
Added an overview of the model as discussed in response to #14. 

Adding more info on the the model specs before they proceed to download
the checkpoints should help folks ensure they have the necessary
resources to effectively utilize Grok-1.
 2024-03-18 09:36:24 -07:00
Szymon Tworkowski
b0e77734fe
Make download instruction more clear (#155) 2024-03-18 09:11:17 -07:00
2 changed files with 28 additions and 1 deletions
Show Stats Expand all files Collapse all files

2
.gitignore vendored Normal file
View File

@ -0,0 +1,2 @@
checkpoints/*
!checkpoints/README.md

27
README.md
View File

@ -2,7 +2,8 @@
This repository contains JAX example code for loading and running the Grok-1 open-weights model.
Make sure to download the checkpoint and place `ckpt-0` directory in `checkpoint`.
Make sure to download the checkpoint and place the `ckpt-0` directory in `checkpoints` - see [Downloading the weights](#downloading-the-weights)
Then, run
```shell
@ -17,13 +18,37 @@ The script loads the checkpoint and samples from the model on a test input.
Due to the large size of the model (314B parameters), a machine with enough GPU memory is required to test the model with the example code.
The implementation of the MoE layer in this repository is not efficient. The implementation was chosen to avoid the need for custom kernels to validate the correctness of the model.
# Model Specifications
Grok-1 is currently designed with the following specifications:
- **Parameters:** 314B
- **Architecture:** Mixture of 8 Experts (MoE)
- **Experts Utilization:** 2 experts used per token
- **Layers:** 64
- **Attention Heads:** 48 for queries, 8 for keys/values
- **Embedding Size:** 6,144
- **Tokenization:** SentencePiece tokenizer with 131,072 tokens
- **Additional Features:**
- Rotary embeddings (RoPE)
- Supports activation sharding and 8-bit quantization
- **Maximum Sequence Length (context):** 8,192 tokens
# Downloading the weights
You can download the weights using a torrent client and this magnet link:
```
magnet:?xt=urn:btih:5f96d43576e3d386c9ba65b883210a393b68210e&tr=https%3A%2F%2Facademictorrents.com%2Fannounce.php&tr=udp%3A%2F%2Ftracker.coppersurfer.tk%3A6969&tr=udp%3A%2F%2Ftracker.opentrackr.org%3A1337%2Fannounce
```
or directly using [HuggingFace 🤗 Hub](https://huggingface.co/xai-org/grok-1):
```
git clone https://github.com/xai-org/grok-1.git && cd grok-1
pip install huggingface_hub[hf_transfer]
huggingface-cli download xai-org/grok-1 --repo-type model --include ckpt-0/* --local-dir checkpoints --local-dir-use-symlinks False
```
# License
The code and associated Grok-1 weights in this release are licensed under the