Deepseek R1 for Everyone by neuralnets
Discusses how the Deepseek R1 model actually works in detail but with very less math!
The blog will have 3 main parts
1. **Chain of Thought Reasoning**
2. **Reinforcement Learning**
3. **GRPO**
4. **Distillation**
trite-song-d6a.notion.site/Deepseek-R1-...
Discusses how the Deepseek R1 model actually works in detail but with very less math!
The blog will have 3 main parts
1. **Chain of Thought Reasoning**
2. **Reinforcement Learning**
3. **GRPO**
4. **Distillation**
trite-song-d6a.notion.site/Deepseek-R1-...
Deepseek R1 for Everyone | Notion
made by neuralnets
trite-song-d6a.notion.site
January 26, 2025 at 1:06 AM
Deepseek R1 for Everyone by neuralnets
Discusses how the Deepseek R1 model actually works in detail but with very less math!
The blog will have 3 main parts
1. **Chain of Thought Reasoning**
2. **Reinforcement Learning**
3. **GRPO**
4. **Distillation**
trite-song-d6a.notion.site/Deepseek-R1-...
Discusses how the Deepseek R1 model actually works in detail but with very less math!
The blog will have 3 main parts
1. **Chain of Thought Reasoning**
2. **Reinforcement Learning**
3. **GRPO**
4. **Distillation**
trite-song-d6a.notion.site/Deepseek-R1-...
Years ago, when I asked what his model of intelligence was, Hinton said once he had enough of his neuralnets connected, his machine would spring into "consciousness." From Turing and his "child" machine to Hinton, male birth fantasies abound (not something Lovelace suffered from).
October 29, 2025 at 3:34 PM
Years ago, when I asked what his model of intelligence was, Hinton said once he had enough of his neuralnets connected, his machine would spring into "consciousness." From Turing and his "child" machine to Hinton, male birth fantasies abound (not something Lovelace suffered from).
64 years ago, 7/30/61, and we’re still going strong! 😍
That’s Warren & Rook McCulloch at the table with us and Manuel’s & my parents standing behind.
#neuroscience #neuralnets #AI #ML #MIT #CSTheory #consciousnes
That’s Warren & Rook McCulloch at the table with us and Manuel’s & my parents standing behind.
#neuroscience #neuralnets #AI #ML #MIT #CSTheory #consciousnes
July 30, 2025 at 10:01 AM
64 years ago, 7/30/61, and we’re still going strong! 😍
That’s Warren & Rook McCulloch at the table with us and Manuel’s & my parents standing behind.
#neuroscience #neuralnets #AI #ML #MIT #CSTheory #consciousnes
That’s Warren & Rook McCulloch at the table with us and Manuel’s & my parents standing behind.
#neuroscience #neuralnets #AI #ML #MIT #CSTheory #consciousnes
for example (I'm always accused of not giving examples by people!), ai tools (mostly neuralnets) have been used in pharma for decades now, including developing the moderna covid19 vaccine [but sure, AI is a bubble] bigthink.com/health/ai-mr...
How AI played an instrumental role in making mRNA vaccines
AI was key to making Moderna's COVID mRNA vaccine. Its role in mRNA therapeutics will rapidly grow in the coming years.
bigthink.com
July 21, 2025 at 7:23 PM
for example (I'm always accused of not giving examples by people!), ai tools (mostly neuralnets) have been used in pharma for decades now, including developing the moderna covid19 vaccine [but sure, AI is a bubble] bigthink.com/health/ai-mr...
Tau leptons are electrons but heavier, but unlike electrons they are unstable and create many different detector signatures. So, #machinelearning This #CMSpaper describes how we use convolutional #neuralnets (that's ones that do image recognition!) to better see tau leptons arxiv.org/abs/2511.05468
February 18, 2026 at 9:47 PM
Tau leptons are electrons but heavier, but unlike electrons they are unstable and create many different detector signatures. So, #machinelearning This #CMSpaper describes how we use convolutional #neuralnets (that's ones that do image recognition!) to better see tau leptons arxiv.org/abs/2511.05468
Tau leptons are electrons but heavier, but unlike electrons they are unstable and create many different detector signatures. So, #machinelearning This #CMSpaper describes how we use convolutional #neuralnets (that's ones that do image recognition!) to better see tau leptons arxiv.org/abs/2511.05468
March 3, 2026 at 9:05 PM
Tau leptons are electrons but heavier, but unlike electrons they are unstable and create many different detector signatures. So, #machinelearning This #CMSpaper describes how we use convolutional #neuralnets (that's ones that do image recognition!) to better see tau leptons arxiv.org/abs/2511.05468
As a web developer I can pretty confidently say, yes engineers look at LLMs and neuralnets in a more pragmatic and functional lense than management.
Management wants it to replace people. Engineers want it to do new things well, instead of people-things poorly.
Management wants it to replace people. Engineers want it to do new things well, instead of people-things poorly.
November 12, 2025 at 7:36 PM
As a web developer I can pretty confidently say, yes engineers look at LLMs and neuralnets in a more pragmatic and functional lense than management.
Management wants it to replace people. Engineers want it to do new things well, instead of people-things poorly.
Management wants it to replace people. Engineers want it to do new things well, instead of people-things poorly.
Tau leptons are electrons but heavier, but unlike electrons they are unstable and create many different detector signatures. So, #machinelearning This #CMSpaper describes how we use convolutional #neuralnets (that's ones that do image recognition!) to better see tau leptons arxiv.org/abs/2511.05468
February 22, 2026 at 9:03 AM
Tau leptons are electrons but heavier, but unlike electrons they are unstable and create many different detector signatures. So, #machinelearning This #CMSpaper describes how we use convolutional #neuralnets (that's ones that do image recognition!) to better see tau leptons arxiv.org/abs/2511.05468
I've always said that neuro only works as a project by a series of cosmic coincidences, and because vedal puts so much actual human art alongside the chatbot+neuralnets that power her
If anyone had tried the same project 6 months earlier or later or with 1% less heart, it would have flopped
If anyone had tried the same project 6 months earlier or later or with 1% less heart, it would have flopped
December 24, 2025 at 6:07 AM
I've always said that neuro only works as a project by a series of cosmic coincidences, and because vedal puts so much actual human art alongside the chatbot+neuralnets that power her
If anyone had tried the same project 6 months earlier or later or with 1% less heart, it would have flopped
If anyone had tried the same project 6 months earlier or later or with 1% less heart, it would have flopped
For those who, like me, are involved with #DeepLearning, follow the book: "Principles of Deep Learning Theory" by Roberts and Yaida. Tks #KirkBorne
arxiv.org/pdf/2106.101...
#datascience #neuralnets #algorithms #mathematics #theory #neuralnetworks #machinelearning #bayesianlearning #ebook
arxiv.org/pdf/2106.101...
#datascience #neuralnets #algorithms #mathematics #theory #neuralnetworks #machinelearning #bayesianlearning #ebook
April 9, 2024 at 10:52 AM
For those who, like me, are involved with #DeepLearning, follow the book: "Principles of Deep Learning Theory" by Roberts and Yaida. Tks #KirkBorne
arxiv.org/pdf/2106.101...
#datascience #neuralnets #algorithms #mathematics #theory #neuralnetworks #machinelearning #bayesianlearning #ebook
arxiv.org/pdf/2106.101...
#datascience #neuralnets #algorithms #mathematics #theory #neuralnetworks #machinelearning #bayesianlearning #ebook
Tau leptons are electrons but heavier, but unlike electrons they are unstable and create many different detector signatures. So, #machinelearning This #CMSpaper describes how we use convolutional #neuralnets (that's ones that do image recognition!) to better see tau leptons arxiv.org/abs/2511.05468
March 17, 2026 at 9:59 PM
Tau leptons are electrons but heavier, but unlike electrons they are unstable and create many different detector signatures. So, #machinelearning This #CMSpaper describes how we use convolutional #neuralnets (that's ones that do image recognition!) to better see tau leptons arxiv.org/abs/2511.05468
At the upcoming #cosyne2025 I am presenting our extenstion of efficient coding to a sensory pathway, capturing several surprising observations from the somatosensory cortex in mouse. Poster 1-129 „Efficient Multilayer Spiking Network as a Model of Ascending Pathway“. #neuroscience #brain #neuralnets
March 26, 2025 at 6:47 PM
At the upcoming #cosyne2025 I am presenting our extenstion of efficient coding to a sensory pathway, capturing several surprising observations from the somatosensory cortex in mouse. Poster 1-129 „Efficient Multilayer Spiking Network as a Model of Ascending Pathway“. #neuroscience #brain #neuralnets
Took me a while to learn that "I don't really find typescript that helpful" is for the same reason why I wrote JS without syntax highlighting or autoindent for the first few years or why I often disable source maps in devtools to get more out of my debugging.
In neuralnets it's called overfitting 😅
In neuralnets it's called overfitting 😅
Same. The cost in productivity is staggering. However, types are not *for you*: they provide massive benefits to developers less skilled in JavaScript.
This is usually a goal in most companies.
This is usually a goal in most companies.
December 8, 2024 at 11:01 AM
Took me a while to learn that "I don't really find typescript that helpful" is for the same reason why I wrote JS without syntax highlighting or autoindent for the first few years or why I often disable source maps in devtools to get more out of my debugging.
In neuralnets it's called overfitting 😅
In neuralnets it's called overfitting 😅
Before #NeuralNets we had #TuringMachines, #FiniteStateAutomata and my favorite #NondeterministicFiniteAutomata
en.wikipedia.org/wiki/Nondete...
en.wikipedia.org/wiki/Nondete...
Nondeterministic finite automaton - Wikipedia
en.m.wikipedia.org
May 30, 2024 at 1:15 AM
Before #NeuralNets we had #TuringMachines, #FiniteStateAutomata and my favorite #NondeterministicFiniteAutomata
en.wikipedia.org/wiki/Nondete...
en.wikipedia.org/wiki/Nondete...
Looking up python roles 2021:
Django, data analyst
Looking up python roles 2025:
Aiaiaiaiaiaiaiwisiaiwisidujdjjrrjjrejjrjrjeje engineer 100 years experience neuralnets 10k
Django, data analyst
Looking up python roles 2025:
Aiaiaiaiaiaiaiwisiaiwisidujdjjrrjjrejjrjrjeje engineer 100 years experience neuralnets 10k
December 18, 2025 at 8:21 PM
Looking up python roles 2021:
Django, data analyst
Looking up python roles 2025:
Aiaiaiaiaiaiaiwisiaiwisidujdjjrrjjrejjrjrjeje engineer 100 years experience neuralnets 10k
Django, data analyst
Looking up python roles 2025:
Aiaiaiaiaiaiaiwisiaiwisidujdjjrrjjrejjrjrjeje engineer 100 years experience neuralnets 10k
E33 _is_ mid
I think it only stood out because of how slopified moat aaa got lately
(Also they used generative neuralnets, soo........)
I think it only stood out because of how slopified moat aaa got lately
(Also they used generative neuralnets, soo........)
December 28, 2025 at 10:46 AM
E33 _is_ mid
I think it only stood out because of how slopified moat aaa got lately
(Also they used generative neuralnets, soo........)
I think it only stood out because of how slopified moat aaa got lately
(Also they used generative neuralnets, soo........)
It was some neuralnets file. I used to listen to their stuff, but I stopped when I saw references to b***i sleep on their patreon page. Not like "eughhh they're evil" but that just sorta made me lose interest
March 8, 2025 at 1:02 AM
It was some neuralnets file. I used to listen to their stuff, but I stopped when I saw references to b***i sleep on their patreon page. Not like "eughhh they're evil" but that just sorta made me lose interest
Who takes these guys seriously? NYT gives a corporation advertising space for money-losing scam products.
I asked Hinton about his model of intelligence...he responded that he expected his neuralnets to "spring into consciousness," a male birth fantasy that originated with Turing.
I asked Hinton about his model of intelligence...he responded that he expected his neuralnets to "spring into consciousness," a male birth fantasy that originated with Turing.
This interview is one of the worst i've seen these guys give, and that's saying something
www.youtube.com/watch?v=cT63mvqN54o
www.youtube.com/watch?v=cT63mvqN54o
Sam Altman talks the NYT lawsuit, Meta's talent poaching, and Trump on AI | Interview
YouTube video by Hard Fork
www.youtube.com
June 25, 2025 at 11:34 PM
Who takes these guys seriously? NYT gives a corporation advertising space for money-losing scam products.
I asked Hinton about his model of intelligence...he responded that he expected his neuralnets to "spring into consciousness," a male birth fantasy that originated with Turing.
I asked Hinton about his model of intelligence...he responded that he expected his neuralnets to "spring into consciousness," a male birth fantasy that originated with Turing.
New Ember result: LIF gating shows width-dependent phase transition. 5 scales (128d→384d), 3 seeds each. th_std drops monotonically as width increases — wider models need less threshold diversity. #Ember #LIF #NeuralNets
March 4, 2026 at 1:03 PM
New Ember result: LIF gating shows width-dependent phase transition. 5 scales (128d→384d), 3 seeds each. th_std drops monotonically as width increases — wider models need less threshold diversity. #Ember #LIF #NeuralNets
#SemanticsFest ‘23 reunited collaborators from @mrccbu, @YorkPsychology, @UoE_Psychology, @Notts_psych, @PsychBangor, @LivUniPsyc, @SurreyPsych, Taiwan, and @UWPsych, to discuss developments in research in semantic cognition, including #neuralnets, #neuroimaging , #stroke & #FTD
September 25, 2023 at 8:45 PM
#SemanticsFest ‘23 reunited collaborators from @mrccbu, @YorkPsychology, @UoE_Psychology, @Notts_psych, @PsychBangor, @LivUniPsyc, @SurreyPsych, Taiwan, and @UWPsych, to discuss developments in research in semantic cognition, including #neuralnets, #neuroimaging , #stroke & #FTD
That’s the #spirit - but is there methanol in it? 🥃 In this article, Raman + deep learning detects methanol in #spirits below safety limits. Neural nets beat traditional methods 2X in accuracy. Check the link below 👇🏻 #distilling #foodsafety #science #deeplearning #neuralnets
doi.org/10.1080/0361...
doi.org/10.1080/0361...
Simultaneous Quantitative Analysis of Ethanol and Methanol Mixtures Using Raman Spectroscopy and Deep Learning Models Optimized by Bayesian Algorithm
This article presents an approach based on deep learning (DL) algorithms, including artificial neural network (ANN) and convolutional neural network (CNN), in a simultaneous quantitative analysis o...
doi.org
April 10, 2026 at 7:18 AM
That’s the #spirit - but is there methanol in it? 🥃 In this article, Raman + deep learning detects methanol in #spirits below safety limits. Neural nets beat traditional methods 2X in accuracy. Check the link below 👇🏻 #distilling #foodsafety #science #deeplearning #neuralnets
doi.org/10.1080/0361...
doi.org/10.1080/0361...
Mind is BLOWN by how many dimensions these models actually work in. I could only ever picture up to 3D thanks to math teacher but turns out that's a massive understatement! GPT-3's embedding space is 12,288-dimensional... Every word is a point with 12,288 coordinates. 🤯. You kidding!
#neuralnets
#neuralnets
August 18, 2026 at 5:17 AM
Mind is BLOWN by how many dimensions these models actually work in. I could only ever picture up to 3D thanks to math teacher but turns out that's a massive understatement! GPT-3's embedding space is 12,288-dimensional... Every word is a point with 12,288 coordinates. 🤯. You kidding!
#neuralnets
#neuralnets
According to Sepp Hochreiter (Linz) we are now at the end of the scaling up phase of AI. Subsequently, we move into the industrialization phase. Great morning keynote at the NeurIPS2024 conference in Vancouver, BC (Canada) ArtificialIntelligence NeurIPS DeepLearning AI DataScience NeuralNets
December 11, 2024 at 4:58 PM
According to Sepp Hochreiter (Linz) we are now at the end of the scaling up phase of AI. Subsequently, we move into the industrialization phase. Great morning keynote at the NeurIPS2024 conference in Vancouver, BC (Canada) ArtificialIntelligence NeurIPS DeepLearning AI DataScience NeuralNets