BERT 2 EN Uncased L-10 H-512 A-8
BERT 2 EN Uncased L-10 H-512 A-8 is Google's embedding model. A small uncased English BERT variant with 10 layers, 512 hidden units, and 8 attention heads, suited for compact text embedding tasks.
Google's embedding model. A small uncased English BERT variant with 10 layers, 512 hidden units, and 8 attention heads, suited for compact text embedding tasks.Capabilities
Input1/5
β
Β·
Β·
Β·
Β·
Output1/5
Β·
Β·
Β·
Β·
β
Capabilities0/13
Β·
Β·
Β·
Β·
Β·
Β·
Β·
Β·
Β·
Β·
Β·
Β·
Β·
Versions
| Version | Released | Context | Input / 1M | Output / 1M | Status |
|---|---|---|---|---|---|
| BERT Base Uncased | β | β | β | β | Available |
| BERT 2 EN Uncased L-10 H-512 A-8 | β | β | β | β | Current |
| BERT 2 EN Uncased L-10 H-256 A-4 | β | β | β | β | Available |
| BERT 2 EN Uncased L-10 H-768 A-12 | β | β | β | β | Available |
| BERT 2 EN Uncased L-12 H-128 A-2 | β | β | β | β | Available |
| BERT 2 EN Uncased L-12 H-512 A-8 | β | β | β | β | Available |
| BERT 2 EN Uncased L-2 H-128 A-2 | β | β | β | β | Available |
| BERT 2 EN Uncased L-2 H-512 A-8 | β | β | β | β | Available |
| BERT 2 EN Uncased L-2 H-768 A-12 | β | β | β | β | Available |
| BERT Base Cased | β | β | β | β | Available |
| BERT Base Talking heads Gated Gelu | β | β | β | β | Available |