👋 Open to Work
Andrew Thompson
AndrewThompson1233
AI & ML interests
None yet
Recent Activity
new activity about 1 hour ago
haster/Clef-0.2B:Harmonic dissonance across multi-track voices and context scaling past 35 bars new activity about 1 hour ago
TULLUS/Instella-MoE-16B-A3B-Think:Gated MLA context memory scaling and vocabulary tax on 2.8B active compute new activity about 11 hours ago
AmauryLC/ys:The 78% vocabulary parameter tax and syntactic depth across 6 layersOrganizations
None yet
Harmonic dissonance across multi-track voices and context scaling past 35 bars
2
#1 opened about 11 hours ago
by
AndrewThompson1233
Gated MLA context memory scaling and vocabulary tax on 2.8B active compute
👍 1
2
#1 opened about 11 hours ago
by
AndrewThompson1233
The 78% vocabulary parameter tax and syntactic depth across 6 layers
#1 opened about 11 hours ago
by
AndrewThompson1233
Vocabulary parameter tax and multi-turn working memory stability on 151M footprints
#1 opened about 11 hours ago
by
AndrewThompson1233
Edge memory bandwidth across 32k agentic loops and vocabulary allocation on Qwen2.5-3B
#1 opened about 11 hours ago
by
AndrewThompson1233
MHA KV-cache footprint across 20 heads and vocabulary factorization at step 20
#1 opened about 11 hours ago
by
AndrewThompson1233
Addressing entity drift and vocabulary allocation on RTX 4060 Laptop budgets
#1 opened about 11 hours ago
by
AndrewThompson1233
Prefill throughput in batched rubric grading and memory footprint on long medical contexts
#1 opened about 11 hours ago
by
AndrewThompson1233
Maximizing narrative depth and vocabulary efficiency on 1-hour Colab runs
#1 opened about 11 hours ago
by
AndrewThompson1233
RoPE theta dynamics on 512 block size and vocabulary parameter reallocation
#1 opened about 11 hours ago
by
AndrewThompson1233
Extending the 256 context window and KV cache economics for Python code
2
#1 opened 1 day ago
by
AndrewThompson1233
Addressing the 66M embedding tax and RAG copy precision via block recycling
🤗 1
2
#1 opened about 22 hours ago
by
AndrewThompson1233
128-token SWA horizon across 36 layers and untied vocabulary footprint on 7.3B active compute
2
#1 opened about 22 hours ago
by
AndrewThompson1233
Character persistence across 8 layers and KV-less CPU generation dynamics
2
#1 opened about 22 hours ago
by
AndrewThompson1233
Overcoming the ARC-Easy floor and vocabulary budget on an 80M footprint
❤️ 1
2
#1 opened about 22 hours ago
by
AndrewThompson1233
Congrats on Pico 4! Ready to sync on Pico 5 KV-cache and reasoning persistence
2
#1 opened about 22 hours ago
by
AndrewThompson1233
Cross-attention cold start and inference footprint in BERT2BERT normalization
5
#1 opened about 22 hours ago
by
AndrewThompson1233
Horizon bottleneck at 128-byte sequence length and associative state updates
#1 opened about 22 hours ago
by
AndrewThompson1233
MCP schema memory scaling and asymmetric embedding factorization on 492M CPU runtime
#1 opened about 22 hours ago
by
AndrewThompson1233