ATOM (atom-ai): code snapshot (100fae0 + batched-train), README with measured numbers, curated checkpoints
9f1076a verified |
Download docs/LIGN_PROMPT_BANK.md from thefinalboss/atom-ai: direct link, hf CLI and curl.
- Browser
- Download file 1.63 kB
-
https://huggingface.co/thefinalboss/atom-ai/resolve/main/docs/LIGN_PROMPT_BANK.md
- Command line
-
hf download hf://thefinalboss/atom-ai/docs/LIGN_PROMPT_BANK.md
-
curl -L -o LIGN_PROMPT_BANK.md https://huggingface.co/thefinalboss/atom-ai/resolve/main/docs/LIGN_PROMPT_BANK.md
1.63 kB
| # L_ign distinct prompt bank — α_bar from fixed prompts | |
| _English. Numbers only. No fluency claim. Generated 2026-09-17 19:22 PDT._ | |
| ## What changed | |
| Flag **`--field-ignorance-prompt-bank`** (default off; **ON** for this run). | |
| Ablate flag **OFF**. | |
| α_bar = mean of detached α snapshots from a fixed probe-style prompt set | |
| (`DEFAULT_IGNORANCE_PROMPTS` in `src/atom_native.py`), refreshed every 256 ticks | |
| with full train-state snapshot/restore. Train α-bank still updates for contrast; | |
| L_ign no longer uses adjacent train ticks for α_bar. | |
| ## Setup | |
| | Item | Value | | |
| |------|-------| | |
| | resume | `atom_native_src.pt` @ **2102500** (field_intel / lign_smoke src) | | |
| | out | `checkpoints/atom_native_lign_prompt_bank/atom_native.pt` @ **2105000** | | |
| | steps | **2500** | | |
| | L_ign | w=**0.08**, m=**0.85**, ablate=**OFF**, prompt-bank=**ON** | | |
| | also | same as prior L_ign smoke (contrast 0.45/0.45, merge ON, slow_every 4) | | |
| | elapsed | **46.5s** (~53.8 tps) | | |
| Artifacts: `docs/artifacts/lign_prompt_bank_2k5/`. | |
| ## Probe | |
| | | before | after | | |
| |--|--------|-------| | |
| | surface logits | **0.992294** | **0.991156** | | |
| | α (field) | **0.720524** | **0.698703** | | |
| ## Train note | |
| **ign → 0.000** after warmup (hinge inactive). Distinct-prompt α_bar makes | |
| `cos(ℓ(α), ℓ(α_bar))` drop **below** margin 0.85 immediately ⇒ **no L_ign gradient**. | |
| mph=**0.500**, merges=**0**. | |
| ## Verdict vs baseline logits ~0.991 | |
| **Fail** on the success metric — after logits **0.991156** (still ~0.991). | |
| α preserved (~**0.698703**). Mechanism side-effect: bank negatives | |
| are “too easy” for the hinge (ign=0), so L_ign stops pushing. | |