2 measurements taken within a single day, net -1. 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 319–320 and does not start at zero.
Measurement log — every subscribers count we have recorded
Measured (UTC)
Subscribers
Change
8 Aug 2026, 02:15
319
-1
7 Aug 2026, 04:48
320
first reading
Engagement
20 posts held, back to 4 March 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.
ERR · 30 days
19.9%
avg views ÷ 319 subscribers
Avg views / post
63.5
2 posts measured
Reaction rate
—
this channel exposes no reaction counts
Posts in window
2
of 20 held
ERR is average views per post over the last 30 days divided by subscribers, the definition TGStat uses, so this figure is comparable with the one you will see elsewhere. It falls structurally as a channel grows: a high ERR on a small channel and a low one on a large channel describe reach mathematics, not quality. We publish the figure and the sample it came from and pass no verdict on it.
ER is defined industry-wide as (forwards + reactions + comments) ÷ views— note the denominator is views, not subscribers. Telegram’s public web preview carries views and reactions but not forward or comment counts, so the reaction rate above is the reactions term only and is therefore a floor: the true ER for this channel is higher by an amount we have not measured and will not estimate.
What these figures were computed from
Window
Rolling 30 days · latest post in window 4 August 2026
Posts held
20 (4 March 2025 – 4 August 2026)
Views total
127
Reactions total
—
Forwards / comments
not exposed by the public surface — not measured, not estimated
Readings taken
7 Aug 2026, 04:48 UTC
Views are a single reading per post, taken at the time above. A post published in the last day or two is still accumulating views, which pulls the 30-day average down slightly. That is a property of the standard definition rather than a fault in it, so we keep the definition rather than “correcting” the number into something nobody can reproduce.
Precision. Telegram publishes view counts on its public widget in short form — 8.12K, 3.7M — so any reading at or above 1,000 reaches us rounded to three significant figures, and only counts below 1,000 are exact. Averages and rates derived from them are shown to the same precision rather than to the unit: a figure like 3,701,250 would assert digits nobody measured.
Reaction counts are published per emoji and rounded the same way, so a total below 1,000 is exact and a larger one is a sum that may carry a rounded component from each emoji above 1,000. Because it is a sum, it does not look rounded — read a large reaction total as three significant figures per contributing emoji rather than as the figure it prints.
Reaction mix
6 reactions across 5 posts, in 4 distinct kinds. The most used accounts for 50.0% of them.
Every reaction kind recorded on the sample, most used first
Reaction
Count
Share
Share, drawn
❤
3
50.0%
👍
1
16.7%
👏
1
16.7%
🔥
1
16.7%
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 5 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 6reactions 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 4 March 2025 to 4 August 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.
Атаки на LLM - это уже не про "сломать чатик"
OWASP в свежем Top 10 для LLM Applications поставил prompt injection на первое место, а рядом отдельно вынес -раскрытие конфиденциальной информации, цепочка поставок и чрезмерное вмешательство - то есть вызовы инструментов, пишет OWASP. - то есть риск давно живёт не только в тексте, но и в том, что модель умеет делать руками через tool-calls.
Самая неприятная часть в том…
"Агенты" буксуют: первый трезвый сигнал с рынка
Руководитель одной из крупных технологических платформ на внутреннем митинге признал, что их AI-агенты "не продвинулись так быстро, как ожидалось": скорость развития и бизнес-эффект от новой структуры с фокусом на агентов не оправдали прогнозы, пишет TechCrunch. При этом компания по‑прежнему рассчитывает увидеть отдачу от AI-инвестиций в горизонте 3–6 месяцев.
Агентски…
🧬 Учёные из Японии создали живую кожу для роботов
Исследователи Токийского университета впервые в истории покрыли лицо робота настоящей кожей, выращенной из человеческих клеток.
🗂Вот ключевые факты об этой инновации:
✅Полный реализм: кожа имитирует текстуру, эластичность и внешний вид настоящей человеческой кожи. Более того, робот может двигать лицевыми мышцами и демонстрировать простейшую мимику — улыбку, нахмуре…
⚠️ Slopsquatting — новая угроза в цепочке поставок ПО, вызванная галлюцинациями ИИ
Если вы когда-либо копировали предложенный ИИ пакет Python и думали: «А это вообще существует?» — вы уже сталкивались с основой атаки под названием slopsquatting.
⚠️ Что это такое?
Slopsquatting — это когда злоумышленники регистрируют несуществующие пакеты, «придуманные» ИИ (например, ChatGPT или GitHub Copilot), надеясь, что кто-то …
🙈Когда ты Data Scientist, но случайно зашёл слишком далеко
🥸В детской книжке по философии обнаружена "диковинка"
Иногда самый сложный вопрос для Data Scientist'ов, с которым он сталкивается — это не «как обучить модель», а «кто теперь хозяин данных?»🤔
🎯 ТОП фреймворков глубокого обучения в 2025 году — что выбрать для проекта?
MXNet - гибкий фреймворк с поддержкой символического и императивного программирования. Отлично масштабируется, идеально подходит для продакшена и тесно интегрирован с AWS. Сложен для новичков, но даёт максимум при работе с большими данными
DL4J (Deeplearning4j) - фреймворк на Java и Scala, отлично интегрируется с Hadoop и Spark. Подходит дл…
🔬 Prompt-in-context-learning: крупнейшая база знаний по работе с LLM от EgoAlpha Lab
📚 Prompt-in-context-learning — это масштабный open-source проект от исследовательской команды EgoAlpha Lab, посвящённый эффективной работе с языковыми моделями. Его цель — собрать воедино лучшие методики prompt engineering и in-context learning, превратив разрозненные подходы в структурированную систему.
🔍 Что вы найдёте в проекте:…
🎮 WHAMM от Microsoft: искусственный интеллект, в который можно играть
Microsoft представила WHAMM (World and Human Action MaskGIT Model) — революционную модель генерации игровых миров, которая позволяет играть прямо внутри нейросети. Это не просто новая нейросетевая архитектура — это демонстрация будущего, в котором границы между реальной игрой и её симуляцией в модели стираются.
🤔 Что делает WHAMM особенной?
🧠 Но…
🚀 NVIDIA G-Assist: Новый AI-помощник для геймеров и создателей контента
NVIDIA представила Project G-Assist — интеллектуального AI-ассистента, который помогает настраивать ПК, оптимизировать игры и управлять системой с помощью голосовых и текстовых команд. Теперь он доступен через NVIDIA App для видеокарт GeForce RTX 30/40/50.
🔥 Что умеет G-Assist?
✅ Оптимизация графики и системы
Автоматически подбирает лучшие наст…
Showing the 12 most recent of 20 posts we hold for @ddtodl. 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.
Polls
The 5 polls we hold for this entry, as Telegram rendered them when we read the post. A poll’s figures keep moving after that, so each one is dated.
У вас есть ограниченный вычислительный ресурс, но необходимо обучить глубокую нейросеть. Какие стратегии вы бы предпочли для хорошего результата?
Использование предварительно обученных моделей и transfer learning69%
Снижение разрядности данных и вычислений (например, mixed precision training)19%
Мини-батчи и градиентное накопление для экономии памяти0%
Адаптивные оптимизаторы (например, AdamW, Ranger) для ускорения сходимости13%
Shares as published, totalling 101%. No per-option vote count is published by Telegram, so none is shown.
Percentages only — there are no per-option vote counts here, because Telegram publishes none.The public post preview gives each option’s share and a single voter total, and nothing else. Multiplying one by the other would produce a per-option tally that looks measured and is not: the shares are rounded to whole numbers before we ever see them. We print what was published and leave the column that does not exist empty.
The shares need not add up to 100.Rounding alone puts many polls at 99 or 101. A poll that allows more than one answer per voter runs well past 100 by design, and several here do. The bars are drawn against a fixed 100% track at each option’s own percentage rather than normalised to the total, so a poll that exceeds it shows that it does instead of being quietly rescaled.
Read from the 20 most recent posts we hold, published 4 March 2025 to 4 August 2026. Telegram labels each poll by kind — an anonymous poll, a quiz, a closed set of final results — and that label is reproduced rather than paraphrased.
Citation-graph rank
Citation-graph rank — 457,183 of 1,169,250entries in the measured graph. A weighted position computed from the forward and mention edges below — republished posts weigh more than named mentions — and recomputed periodically, over the whole graph. Published only as this ordinal position, never as a score: a position is a fact, and a score printed beside one channel’s name would read as a verdict this register does not make. The two counts beneath stay separate for the same reason mentions are never summed with forwards anywhere else on this page — a named-by count costs nothing to manufacture. The top 100 by this measure, or how it is computed.
Forward network
Republished by
Channels on the register that have forwarded this channel's posts into their own feed.
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.
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 8 August 2026 — this
entry's latest reading, not the date you are reading this.
“Deep Dive 2 Deep Learning” (@ddtodl), 319 subscribers as measured 8 August 2026. Telegram Register, tgregister.com/channel/ddtodl.
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.
