Downloads · 30 days
58
2% of all-time downloads
alon-albalak/ODM_1Bparams_50Btokens_seqlen2048
ODM_1Bparams_50Btokens_seqlen2048 is a feature extraction model from alon-albalak. Use it when you need embeddings to search or compare text. It is set up for transformers. The card lists the license as openrail.
This is a GPT-NeoX model trained on 50 billion tokens from The Pile, using the Online Data Mixing method.
Downloads · 30 days
58
2% of all-time downloads
All-time downloads
3.9K
Public
Parameters
976M
7.4 GB on disk
Likes
0
Public
Click a slice to open those files.
.bin3.7 GB · 50%
How the weights are stored.
F32909M · 93%
From the Hugging Face model README
This is a GPT-NeoX model trained on 50 billion tokens from The Pile, using the Online Data Mixing method.
The OpenLLM leaderboard won't let me submit my model because the description is too short, so I'm adding more characters to the description in hopes that it will be evaluated.