Transformers grown layer-by-layer on frozen embeddings. Explores emergent capabilities with depth.
Andrey
Bochkov
AI & ML interests
None yet
Recent Activity
authored
a paper
1 day ago
Emergent Semantics Beyond Token Embeddings: Transformer LMs with Frozen
Visual Unicode Representations
updated
a collection
1 day ago
Tokenizers
updated
a collection
1 day ago
Pro models [pretrain]
Organizations
None yet
models
25
Bochkov/bvv241-2-3
Updated
•
13
Bochkov/bvv241-max
Updated
•
12
Bochkov/bvv241-nemo
Updated
•
7
Bochkov/bvv241-abs
Updated
•
12
Bochkov/max_bvv_ru
Updated
•
23
Bochkov/max_bvv_zh
Updated
•
20
Bochkov/max_bvv_moe
Updated
•
15
Bochkov/nemo_bvv_ru
Updated
•
16
Bochkov/nemo_bvv_zh
Updated
•
15
Bochkov/nemo_bvv_moe
Updated
•
21
datasets
0
None public yet