mmBERT is trained on 3T tokens from over 1800 languages, showing SoTA scores on benchmarks and exceptional low-resource performance
AI & ML interests
None defined yet.
Recent Activity
Papers
Genomic Next-Token Predictors are In-Context Learners
Controlled Generation for Private Synthetic Text
models 53
jhu-clsp/mmBERT-small
Fill-Mask • Updated
• 22.2k • • 64
jhu-clsp/mmBERT-base
Fill-Mask • Updated
• 277k • • 193
jhu-clsp/mmBERT-checkpoints
Updated
• 4
jhu-clsp/ettin-decoder-1b
Fill-Mask • Updated
• 186 • 5
jhu-clsp/ettin-decoder-32m
Text Generation • Updated
• 238
jhu-clsp/ettin-encoder-1b
Feature Extraction • Updated
• 737 • 21
jhu-clsp/ettin-encoder-68m
Fill-Mask • Updated
• 9.97k • 4
jhu-clsp/ettin-dec-from-enc-32m
Text Generation • Updated
• 4
jhu-clsp/ettin-encoder-150m
Fill-Mask • Updated
• 23.2k • • 9
jhu-clsp/ettin-decoder-400m
Text Generation • Updated
• 158 • 4
datasets 38
jhu-clsp/megawika-2
Updated
• 131 • 2
jhu-clsp/mmBERT-decay-data
Updated
• 11.7k • 5
jhu-clsp/mmBERT-midtraining-data
Updated
• 16.4k • 1
jhu-clsp/ettin-pretraining-data
Updated
• 4.68k • 8
jhu-clsp/ettin-decay-data
Updated
• 1.5k • 1
jhu-clsp/astro-llms-benchmark-dataset
Viewer
• Updated
• 40 • 109
jhu-clsp/astro-llms-full-query-data
Viewer
• Updated
• 368 • 99
jhu-clsp/ettin-extension-data
Updated
• 1.48k
jhu-clsp/ettin-data-order
Viewer
• Updated
• 3B • 5 • 1
jhu-clsp/rank1-R1-MSMARCO
Viewer
• Updated
• 635k • 40 • 2