AMajor/grok-1
1
0
Fork 0
mirror of https://github.com/xai-org/grok-1.git synced 2024-11-23 12:09:52 +03:00
Code Issues Packages Projects Releases Wiki Activity

Compare commits

base: AMajor:e5ea38bde9d6e517b0e266f79245dabe028bcd25
Branches Tags
AMajor:main
AMajor:download-instruction
...
compare: AMajor:929a1d2ce85ff921a59efb68a419e988ea698101
Branches Tags
AMajor:main
AMajor:download-instruction

3 Commits
e5ea38bde9 ... 929a1d2ce8

Author SHA1 Message Date
Andrew Kean Gao
929a1d2ce8
Merge 4e2e30bd6f into 7050ed204b 2024-03-19 08:54:27 -07:00
Eddy
7050ed204b
Corrected name of package "cuda12-pip" (#194) 
The `cuda12-pip` package was wrongly named `cuda12_pip`
in requirements.txt
 2024-03-19 08:48:22 -07:00
Andrew Kean Gao
4e2e30bd6f
Update README.md to have context length higher up 
In my original summary of the model specifications, I had put the context length near the bottom, but upon thought, it is probably one of the relevant details to end-users, so it should be higher.

Also, "Additional Features" should be the final bullet point for editorial reasons.
 2024-03-18 10:43:55 -07:00
2 changed files with 2 additions and 2 deletions
Show Stats Expand all files Collapse all files

2
README.md
View File

@ -25,6 +25,7 @@ Grok-1 is currently designed with the following specifications:
- **Parameters:** 314B - **Parameters:** 314B
- **Architecture:** Mixture of 8 Experts (MoE) - **Architecture:** Mixture of 8 Experts (MoE)
- **Experts Utilization:** 2 experts used per token - **Experts Utilization:** 2 experts used per token
- **Maximum Sequence Length (context):** 8,192 tokens
- **Layers:** 64 - **Layers:** 64
- **Attention Heads:** 48 for queries, 8 for keys/values - **Attention Heads:** 48 for queries, 8 for keys/values
- **Embedding Size:** 6,144 - **Embedding Size:** 6,144
@ -32,7 +33,6 @@ Grok-1 is currently designed with the following specifications:
- **Additional Features:** - **Additional Features:**
- Rotary embeddings (RoPE) - Rotary embeddings (RoPE)
- Supports activation sharding and 8-bit quantization - Supports activation sharding and 8-bit quantization
- **Maximum Sequence Length (context):** 8,192 tokens
# Downloading the weights # Downloading the weights

2
requirements.txt
View File

@ -1,4 +1,4 @@
dm_haiku==0.0.12 dm_haiku==0.0.12
jax[cuda12_pip]==0.4.25 -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html jax[cuda12-pip]==0.4.25 -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html
numpy==1.26.4 numpy==1.26.4
sentencepiece==0.2.0 sentencepiece==0.2.0