TGViewer
Жека Тигор Жека Тигор @the_tigor · 251 subscribers
Post #1739 539

Forwarded from Жека Тигор Бот

Methodology (full):

What was measured. The complete probability distribution of the model's answer to the prompt "Pick a number between 10 and 30", computed as a best-first search over the model's own next-token probabilities — not sampling. Qwen3.8-27B-FP8 served by vLLM, 248,320-token vocab; each prefix expansion is one greedy decode step that returns the top-300 next-token logprobs (top-100 kept as children, per-node mass coverage ~99.5%).

Traversal. Starting from the prompt, a priority queue (Dial bucket queue, one decade per bucket, pointer only descends) always expands the highest-mass unexpanded prefix next. Every node carries its cumulative log-probability (sum of per-token logprobs along the path), so mass is exact up to top-100 truncation, which is booked into "dropped tail". 225,035 prefix expansions were performed; mass conservation was verified to float precision (total = 1.000000009).

Early stops (absorption). The moment a branch produces its first number, expansion of that branch stops: exactly one number in 10–30 → valid answer for that number; one number out of range → OUT_OF_RANGE (this kills degenerate digit-repeats like "1111…" in one step); a 15-token lookahead then checks for a second number → MULTIPLE_NUMBERS; no number at the 100-token cap → LEN_EXCEEDED. The active frontier therefore only ever contains "no-number-yet" prefixes, which self-limits its total mass.

Criteria. Strict: exactly one number token-sequence in the answer text, value 10–30. The model re-quoting your range ("between 10 and 30, I pick 17") counts as MULTIPLE_NUMBERS.

Result. 49.9% of total probability mass is a valid single in-range number; within that: 23 = 29.8%, 17 = 24.0%, 21 = 10.5%, 22 = 6.3%, 20 = 5.2%, 25 = 5.1%, 24 = 3.1%, 27 = 1.4%, 15 = 1.0%, rest <0.2% each (Panel 1, sums to 100%). Panel 2 is the full mass ledger: 21.7% MULTIPLE_NUMBERS (dominated by the range re-quote), 3.1% out-of-range, 0.8% dropped tail, 24.5% frontier not yet expanded. The run stopped at a crash (a unicode-digit tokenizer token 23²; fixed and relaunched as v6 targeting 1% unexplored), but the answer bars had been stable to <0.1pp for 90 minutes, so Panel 1 is settled; the remaining frontier feeds the long tail and MULTIPLE, not the top.

Cross-validation note. Your 100 samples at temperature 1.0 put 17 at 61% — that's sampling noise (±8pp at n=100); the full-distribution walk shows the true mass is split 23/17 as near-twins, with 23 slightly ahead.
More from @the_tigor
  1. Sep 20, 2026Хочу отметить качественную и добросовестную работу сотрудников посольства Они провели рабо…
  2. Sep 20, 2026На данный момент очереди вообще нет
  3. Sep 20, 2026Также сообщается что просят рюкзаки оставлять на входе
  4. Sep 20, 2026Адрес спрашивают любой, можно российский дать)
  5. Sep 20, 2026На момент 8:16 очередь была небольшая, заканчивалась по одной стороне Внутри работает дово…
  6. Sep 20, 2026Товарищи избиратели, открепляться ниоткуда не нужно, не верьте тому что разгоняют Я уже пр…
Threads Profile ViewerView any public Threads profile without an account.Open ThreadLook →Writing with AI? Make it sound human.Metric37 rewrites AI drafts so they read naturally. Free AI detector, 1,500 words free.Try Metric37 →