epimem: One-shot gradient-free learning on frozen transformers

Browse Source

- paper.md: full paper (Tommi Niemi / Rotko Networks)
- python/epimem.py: standalone Python reproduction
- export_onnx.py: ONNX export from HuggingFace (generates model files)
- results/memory_bank.json: example hidden-state vectors (896-dim)
- schema/: FlatBuffer schemas for memory bank + organism
- models/tokenizer/: Qwen 2.5 tokenizer files

Run: pip install transformers torch && python python/epimem.py
(Downloads Qwen 2.5 automatically from HuggingFace)

...

This commit is contained in:

Tommi Niemi

2026-04-05 01:42:13 +07:00

parent c26496dacf

commit 404df92ade

12 changed files with 785695 additions and 2 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download Patch File Download Diff File Expand all files Collapse all files

2753

results/memory_bank.json Normal file

File diff suppressed because it is too large Load Diff