🔄 In a Training Loop

Andrey

Bochkov

·

AI & ML interests

None yet

Organizations

None yet

Articles 1

Article

1

Emergent Semantics Beyond Token Embeddings: A GPT-like Transformer Learns with Frozen 16‑D Binary Token-ID Embeddings (n_embed=16)

Collections 4

View 4 collections

Papers 2

arxiv:2507.07129

arxiv:2507.04886

models 19

Bochkov/llm-fix-min-affine-recoded-minimal-code-table-free

Text Generation • 0.5B • Updated May 12 • 24

Bochkov/llm-fix-min-fixed-minimal-binary-code

Text Generation • 0.5B • Updated May 12 • 21

Bochkov/llm-fix-min-baseline-learned-input-table-model-classic

Text Generation • 0.5B • Updated May 12 • 16

Bochkov/growing-transformers-model-frozen-16-bit-baseline-monolyth-181m

Text Generation • 0.2B • Updated Jan 9 • 20

Bochkov/growing-transformers-model-unfrozen-baseline-monolyth-247m

Text Generation • 0.2B • Updated Jan 9 • 14

Bochkov/growing-transformers-model-unfrozen-1-9-247m

Text Generation • 0.2B • Updated Jan 9 • 80

Bochkov/growing-transformers-model-16-bit-1-9-181m

Text Generation • 0.2B • Updated Jan 9 • 91

Bochkov/growing-transformers-model-frozen-unicode-baseline-monolyth-247m

Text Generation • 0.2B • Updated Jan 9 • 13

Bochkov/emergent-semantics-model-uni-glyph-335m

Text Generation • 0.3B • Updated Jan 7 • 15

Bochkov/emergent-semantics-model-unfrozen-335m

Text Generation • 0.3B • Updated Jan 7 • 12

datasets 0

None public yet