kent/consciousness
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
consciousness/training
History
ProofOfConcept c5d7d8cb5d apollo-mini training system: initial implementation 
Core components for online fine-tuning of Qwen3.5-27B with CUDA IPC
shared weight memory between vLLM and the training process:

- apollo_mini.py: rank-1 optimizer (SGD memory, AdamW quality)
- apollo_worker.py: HTTP daemon coordinating training with vLLM
- weight_mapping.py: vLLM merged → HF separate layout (zero-copy views)
- training_example.py: tokenization with chat template
- export_weights.py: CUDA IPC handle export from vLLM
- train.py: standalone training script (alternative to daemon)
- DESIGN.md: architecture and protocol documentation

Validated: CUDA IPC autograd works on real Qwen3.5 weights (B200).
Apollo-Mini rank-1 projection + scaling + in-place update confirmed.

Co-Authored-By: Kent Overstreet <kent.overstreet@gmail.com>
2026-03-30 22:02:37 -04:00
..
apollo_mini.py apollo-mini training system: initial implementation 2026-03-30 22:02:37 -04:00
apollo_worker.py apollo-mini training system: initial implementation 2026-03-30 22:02:37 -04:00
DESIGN.md apollo-mini training system: initial implementation 2026-03-30 22:02:37 -04:00
export_weights.py apollo-mini training system: initial implementation 2026-03-30 22:02:37 -04:00
train.py apollo-mini training system: initial implementation 2026-03-30 22:02:37 -04:00
training_example.py apollo-mini training system: initial implementation 2026-03-30 22:02:37 -04:00
weight_mapping.py apollo-mini training system: initial implementation 2026-03-30 22:02:37 -04:00