Channel

Subjective Intelligence

@subj_int

On this record: Growth · Engagement · What this channel posts · Reactions · Posts · Citations · Cite this entry

86subscribers

+0 since we began measuring on 7 August 2026

Risers and fallers across the register · movement among entries of Under 1,000.

Register entry

Telegram ID	-1001696489856
Type	Channel
Username	@subj_int
Description	Искусственный интеллект на широкую ногу, и как пойдет.
Created	Between 1 December 2021 and 31 March 2023— estimated from Telegram’s id allocation, not measured. How this range is calculated.
First recorded	9 August 2026
Last confirmed live	9 August 2026
Measurements held	2
On Telegram	t.me/subj_int

Growth

2 measurements spanning 2 days. Dots are measurements; the straight line between them is drawn to join them, not to claim we know the path taken in between — snapshots are recorded only when a count changes, so gaps mean “no change observed”, never “interpolated”. The vertical axis spans 85–87 and does not start at zero.

Measurement log — every subscribers count we have recorded
Measured (UTC)	Subscribers	Change
9 Aug 2026, 16:31	86	no change
7 Aug 2026, 13:17	86	first reading

Engagement

20 posts held, back to 5 June 2025 — the reader has not yet reached the start of this channel’s public history, so older posts may sit further back, unread. Read across 1 pageof Telegram’s post history, 20 posts per page.

Nothing published in the last 30 days. ERR and ER are rolling 30-day measures, so there is nothing to compute — we hold 20 posts for this entry, the most recent from 24 February 2026. An engagement rate over an empty window would be a number about nothing.

What this channel posts

Photos: 29
Videos: 4
Links: 124

Lifetime counters from Telegram’s own channel header, read 9 August 2026 — not the date at the top of this page, which is when the subscriber count was last read. Below Telegram’s rounding threshold, so these counts are exact.

Reaction mix

42 reactions across 18 posts, in 4 distinct kinds. The most used accounts for 57.1% of them.

Every reaction kind recorded on the sample, most used first
Reaction	Count	Share
❤	24	57.1%
🔥	10	23.8%
👍	6	14.3%
🤯	2	4.76%

No sentiment is inferred, and none should be read in. This table is ordered by count and by nothing else. Emoji do not carry stable meaning across languages or communities — 🙏 is thanks in one channel and mourning in another — so we publish which ones were pressed and how often, and pass no judgement on what an audience meant by them.

Precision. Telegram publishes reaction counts per emoji and short-forms each one — 4.34K, 1.2M — so any single kind at or above 1,000 reaches us at three significant figures, and only counts below 1,000 are exact. The shares above are ratios of those figures and carry the same error. This is also why the total here can differ slightly from a reaction total printed elsewhere on the page: both are sums of the same rounded parts, taken over samples with different edges.

Coverage. Reactions were read on 19 of the 20 sampled posts in this sample. Summed by Telegram’s own count on each post — not by adding up the per-emoji breakdown above — those same posts carry 42reactions in total: the kind of figure the paragraph above means by “a reaction total printed elsewhere on the page”.

Measured over the 20 most recent posts we hold, published 5 June 2025 to 24 February 2026, using the newest reading held for each. Telegram Stars are excluded: they are a payment, not a reaction, and they have their own section.

Recent posts

24 Feb 2026, 17:54 UTC119 views2 reactionsread 9 August 2026

👻 A Comprehensive Survey of Self-Evolving AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systems paper github

❤1👍1

23 Feb 2026, 13:15 UTC124 views1 reactionsread 9 August 2026

🍒 Lenny’s Podcast. Head of Claude Code: What happens after coding is solved | Boris Cherny - Head of Claude Code: no manual coding since Nov 25 - 10-20-30 PRs per day - Coding is largely solved - Don’t cut/limit tokens budget too early - Build for latent demand (eg. Claude Code -> Claude Cowork) - Fast iteration is a key (fix in minutes not weeks) - Build for model 6 months from now - Don’t waste time on extensive s…

❤1

22 Feb 2026, 11:16 UTC150 views2 reactionsread 9 August 2026

По ИИ-канальчикам пролетела ссылка с резюме текущих техник тренировки больших моделей от Alex Wa. Возможно, она не включает какие-то особо секретные подходы крупных лабораторий, но в целом дает очень хороший обзор, что надо знать, чтобы обучить фронтирную модель: frontier model training methodologies.

❤2

22 Feb 2026, 11:03 UTC128 views3 reactionsread 9 August 2026

Дальше - спекуляции про будущее, хоть и ближайшее, которые в основном дополняю от своего имени. Что уже существует в прототипах, и явно будет масштабироваться: 🌺 2026-2027: Технологии: 👉 Многоагентные системы (MAS, agent swarms). Уже есть работающие прототипы, которые декомпозируют исходную задачу на десятки и сотни подзадач, решаемых параллельно подпроцессами/субагентами с разной специализацией. Например, такие ро…

❤3

22 Feb 2026, 11:02 UTC106 views1 reactionsread 9 August 2026

Гениальный популяризатор фронтира для смертных Sebastian Raschka в своем обзоре The State Of LLMs 2025: Progress, Problems, and Predictions (ютуб версия здесь) сформулировал, как менялся фокус фронтирных исследований по годам ("эпохи/эры LLM"). Я его прокомментирую и немного дополню/расширю. 🌺 The LLM Eras: 👉 202x Pre-training (foundation) - работала базовая идея скейлинга сетей и датасетов, искались модификации ар…

🔥1

11 Feb 2026, 06:59 UTC107 views5 reactionsread 9 August 2026

Если вы искали у кого поучиться, то вот эти парни знают, что делают: THE METIS LIST The World’s Top AI Researches У всех них есть google scholar, twitter и множество интервью на ютубе, из которых можно немного понять, как и о чем они думают, и что нас ждет в ближайшем будущем.

👍3❤2

31 Jan 2026, 15:15 UTC102 views2 reactionsread 9 August 2026

🧠 State of Brain Emulation 2025: Research Report Niccolo Zanichelli, Maximilian Schons, Isaak Freeman, Philip K. Shiu, Anton Arkhipov Comprehensive 2025 report on whole brain emulation: 20 years of progress in neural recording, connectomics, and computational neuroscience.

❤2

25 Dec 2025, 13:19 UTC149 views3 reactionsread 9 August 2026

🍎 Ризонинг — изначально не сильная сторона LLM, по крайней мере в пространстве токенов. Кажется, нашелся подходящий слой, куда и как его правильно вынести, а LLM-кам можно оставить семантику. Описанное выше — промышленный стандарт на ближайшее время. 🍐 Альтернативная ветка развития — ризонинг в латентном пространстве. Это, например, архитектуры JEPA Лекуна, ради исследования которых он ушел из Meta, и много разных д…

🔥3

25 Dec 2025, 13:19 UTC129 views3 reactionsread 9 August 2026

Итак, не считая сплошного, почти ежедневного потока различных оптимизаций (стоимость продакшн инференса падает в тысячи раз), новый state of the art выглядит так: 👉 1. Прирост эффектов от скейлинга стал замедляться, плюс таки кончились тексты в интернете, зато стало общепризнанным фактом, что дообучение моделей на цепочках рассуждений и примерах промежуточных шагов хорошо работает - между ними проще устанавливаются …

🔥3

5 Sept 2025, 18:44 UTC89 views2 reactionsread 9 August 2026

Красивая постановка вопроса. Какими минимальными изменениями возможно перейти от обучения «фрагментированных запутанных представлений» к обучению «унифицированных факторизованных представлений»? : https://t.me/gonzo_ML_podcasts/798

❤2

5 Aug 2025, 14:36 UTC152 views3 reactionsread 9 August 2026

☀️ Genie 3 от Google Deepmind Впечатляет. Сначала были первые генеративные картиночные модели, потом видеомодели, потом появились интерактивные модели игровых миров, и вот теперь следующее слово в world models - одном из ключевых компонентов для развитого восприятия и планирования. Genie 3: - генерирует анимированный интерактивный мир по текстовому промпту - в высоком качестве кадра (720p) и в реальном времени (24 f…

👍2🤯1

1 Aug 2025, 15:04 UTC88 views4 reactionsread 9 August 2026

Forwarded from @neurocampus2030Photo

🌞Остался последний месяц лета! Предлагаем вам провести его с пользой и погрузиться в подборку курсов в области нейронаук: ⇥ «Основы нейробиологии», Филипп Хайтович ⇥ «Основы нейровизуализации, нейроинтерфейсы», Алекс Осадчий, Алексей Тумялис ⇥ «Введение в неврологию и нейрохирургию», Михаил Синкин и Вадим Русскин ⇥ «Введение в молекулярную биологию», Дарья Серебряная ⇥ «Теория эволюции нервной системы», Дарья Р…

❤2🔥2

Showing the 12 most recent of 20 posts we hold for @subj_int. View and reaction counts are the latest single reading for each post, not a live figure, and a recent post is still accumulating both. A view count marked ≈ was rounded by Telegram before we ever saw it — t.me prints views in full below 1,000 and to three significant figures above, so ≈1,200,000 means somewhere between 1,150,000 and 1,249,999. Unmarked counts are exact. Text is reproduced from the public post preview and truncated for length.

Forward network

Republishes

Channels on the register whose posts this channel has forwarded.

Нейрокампус
@neurocampus2030 · 4,0341 post

Built only from forwarded posts we have actually read, on both sides. Coverage is early and deliberately incomplete: a missing link means we have not read the post that would prove it, never that the relationship does not exist. Counts are distinct forwarded posts observed, so they only ever go up as we read more.

Mentions

Names

Channels on the register whose handles appear in this channel's posts.

gonzo_ML_podcasts
@gonzo_ML_podcasts · 3,0773 posts эйай ньюз
@ai_newz · 95,8921 post LLM под капотом
@llm_under_hood · 28,6541 post Нейрокампус
@neurocampus2030 · 4,0341 post red_mad_robot
@Redmadnews · 11,4461 post Kali Novskaya
@rybolos_channel · 18,9991 post

A mention is a weaker signal than a forward and is counted separately for that reason — naming a channel is not republishing it, and a handle in a post body is easy to place deliberately. The post counts beside each row below are distinct posts in which the handle appeared, from posts we have read on both sides — the “Named by N registered channels” figure above is a different count, of distinct NAMING CHANNELS rather than posts, and is not the sum of the rows under it.

Cite this entry

A live page changes as we take new readings, so a citation should name the measurement it is based on, not just the URL. The line below cites the subscriber count as measured 9 August 2026 — this entry's latest reading, not the date you are reading this.

“Subjective Intelligence” (@subj_int), 86 subscribers as measured 9 August 2026. Telegram Register, tgregister.com/channel/subj_int.

Full measurement history, CC BY 4.0. Every reading this register holds for this entry, not just the latest one, as a dated, downloadable record: CSV · JSON. Free to use with attribution to tgregister.com. Each file carries its own generation timestamp, which is the figure to cite for exactly when the data was retrieved.