wassname and Claudypoo
69d7d2f1d9
demo: apply gpt-5.5 review -- float-C fix, trim comments, soften overclaims
...
External review (docs/reviews/code_demo.md) triaged scout-mindset:
- FIX float-C crash: C={C:+g} not {:+d} (steering coeffs are floats)
- ACCEPT trim: shorter demo.py/config/fit docstrings (user also flagged verbosity)
- ACCEPT soften "fit J where we steer" -> "closer to the chat distribution" (most
fitted positions are user/doc tokens, not assistant <think>; run-524 went further)
- ACCEPT soften "what the model thinks" -> "lens readout (linear approx)"
- ADD seed to show_steer so per-C blocks are comparable under sampling
- REJECT "</think> stripped by skip_special_tokens" -- verified false: decode keeps
think tags (they're added, not registered-special tokens), split_think works
Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com >
2026-07-10 15:02:42 +08:00
wassname
474f74ac33
wip
2026-07-10 14:36:55 +08:00
wassname and Claudypoo
ebb912452d
U4 step3: dim_batch 16->8 to survive OOM contention with user's VS Code kernel
...
Run 551 was OOM-killed at n_done=36: the user's VS Code Jupyter kernel
(jsteer venv, PID 3214401) co-loaded ~1.5GB VRAM + 1.9GB RAM while the fit
sat at the 22.4/24.6GB ceiling. Clean SIGKILL with no CUDA traceback = host
OOM killer, not a CUDA OOM. dim_batch=8 halves the fit's peak footprint;
it changes only the backward schedule, not the accumulated Jacobian, so U4
exactness is preserved. Resumes from checkpoint (n_done=36), lossless.
Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com >
2026-07-10 13:56:17 +08:00
wassname and Claudypoo
4fb1cfd63e
apply external review (deepseek): reject float bands post-fit, assert->ValueError, 2 clarifying comments
...
Review verdict APPROVE; rejected findings (empty-prompts guard = preemptive
defensive check, from_hf dedup = ms-scale, lm.forward swap = loses attention
mask on padded batches) documented in docs/reviews/code.md triage.
Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com >
2026-07-10 13:05:27 +08:00
