Knicks Opponents Points
ATL1 102
ATL2 107
ATL3 109
ATL4 98
ATL5 97
ATL6 89
PHI1 98
PHI2 102
PHI3 94
That’s elite shit.
ATL1 102
ATL2 107
ATL3 109
ATL4 98
ATL5 97
ATL6 89
PHI1 98
PHI2 102
PHI3 94
That’s elite shit.
May 9, 2026 at 3:25 PM
Knicks Opponents Points
ATL1 102
ATL2 107
ATL3 109
ATL4 98
ATL5 97
ATL6 89
PHI1 98
PHI2 102
PHI3 94
That’s elite shit.
ATL1 102
ATL2 107
ATL3 109
ATL4 98
ATL5 97
ATL6 89
PHI1 98
PHI2 102
PHI3 94
That’s elite shit.
Might do Feynman diagrams earlier, idk. Wicks theorem and like baby’s first Feynman diagrams (like 4th moment of a Gaussian) and then phi3+phi4 after OM path integral maybe.
Today I’ll cover the Ito Lemma!
Next week I’ll do “What is a Markov Process”
Then the OM path integral
Then MSRDJ path integral
Then Fokker-Planck and Ornstein Uhlenbeck process
Then Feynman diagrams!
Next week I’ll do “What is a Markov Process”
Then the OM path integral
Then MSRDJ path integral
Then Fokker-Planck and Ornstein Uhlenbeck process
Then Feynman diagrams!
February 6, 2026 at 12:30 AM
Might do Feynman diagrams earlier, idk. Wicks theorem and like baby’s first Feynman diagrams (like 4th moment of a Gaussian) and then phi3+phi4 after OM path integral maybe.
New paper by @andrewpiper.bsky.social et al. makes several contributions to the study of social networks in fiction. 1. A new dataset of character interactions (n = 13,395) 2. A small language model (Phi3-7B), trained to label character interactions. 3. Also some proof of concept results. #MLSky
panoptes-uploads.zooniverse.org
October 20, 2024 at 5:55 PM
New paper by @andrewpiper.bsky.social et al. makes several contributions to the study of social networks in fiction. 1. A new dataset of character interactions (n = 13,395) 2. A small language model (Phi3-7B), trained to label character interactions. 3. Also some proof of concept results. #MLSky
should have used a local phi3 mini
June 24, 2024 at 3:26 PM
should have used a local phi3 mini
Have been using Mistral:latest for a long time both as a llamafile and via ollama, still works really well for me. (Have also been trying out all of these variations for vibe checks recently)
December 29, 2024 at 1:21 AM
Have been using Mistral:latest for a long time both as a llamafile and via ollama, still works really well for me. (Have also been trying out all of these variations for vibe checks recently)
我有个想法 给各平台的私信都写个小程序 直接把私信接到Phi3里来抵（戏）御（耍）骚扰者🤔
January 15, 2025 at 12:31 PM
我有个想法 给各平台的私信都写个小程序 直接把私信接到Phi3里来抵（戏）御（耍）骚扰者🤔
Cómo ejecutar llama3, phi3 en tu PC local utilizando Ollama - Planeta Chatbot https://planetachatbot.com/como-ejecutar-llama3-phi3-en-pc-local-utilizando-ollama/
Cómo ejecutar llama3, phi3 en tu PC local utilizando Ollama - Planeta Chatbot
planetachatbot.com
December 15, 2024 at 5:44 PM
Cómo ejecutar llama3, phi3 en tu PC local utilizando Ollama - Planeta Chatbot https://planetachatbot.com/como-ejecutar-llama3-phi3-en-pc-local-utilizando-ollama/
Okay. New plan. This is the program counter. It's a 16-bit binary counter, with tri-state buffers on the outputs. On PHI1, PHI2, and PHI3 rising edge, it increments the counter and opens the buffers, putting an address on the bus each time.
December 1, 2025 at 11:58 PM
Okay. New plan. This is the program counter. It's a 16-bit binary counter, with tri-state buffers on the outputs. On PHI1, PHI2, and PHI3 rising edge, it increments the counter and opens the buffers, putting an address on the bus each time.
If this approach makes Phi3-mini-3.8B better at math than o1-preview, what happens when the same thing is done with an o1/o3-class model?
January 11, 2025 at 2:00 PM
If this approach makes Phi3-mini-3.8B better at math than o1-preview, what happens when the same thing is done with an o1/o3-class model?
Sécurité routière : + 15 % de tués en plus sur les routes des Landes
🔗https://l.sudouest.fr/Phi3
🔗https://l.sudouest.fr/Phi3
November 4, 2025 at 7:24 PM
Sécurité routière : + 15 % de tués en plus sur les routes des Landes
🔗https://l.sudouest.fr/Phi3
🔗https://l.sudouest.fr/Phi3
Given you can already download ALL of wikipedia locally, I don't see why someone hasn't already done this. It seems likely a small language model like Phi3 could make an excellent front end. Is it the training cost?
November 4, 2025 at 7:50 PM
Given you can already download ALL of wikipedia locally, I don't see why someone hasn't already done this. It seems likely a small language model like Phi3 could make an excellent front end. Is it the training cost?
Microsoft releases rStar-Math: "Small LLMs can master math reasoning with self-evolved deep thinking"
It improves Qwen2.5-Math-7B from 58.8% to 90.0% and Phi3-mini-3.8B from 41.4% to 86.4%, surpassing o1-preview by +4.5% and +0.9%
Source : arxiv.org/html/2501.04...
It improves Qwen2.5-Math-7B from 58.8% to 90.0% and Phi3-mini-3.8B from 41.4% to 86.4%, surpassing o1-preview by +4.5% and +0.9%
Source : arxiv.org/html/2501.04...
January 9, 2025 at 9:19 AM
Microsoft releases rStar-Math: "Small LLMs can master math reasoning with self-evolved deep thinking"
It improves Qwen2.5-Math-7B from 58.8% to 90.0% and Phi3-mini-3.8B from 41.4% to 86.4%, surpassing o1-preview by +4.5% and +0.9%
Source : arxiv.org/html/2501.04...
It improves Qwen2.5-Math-7B from 58.8% to 90.0% and Phi3-mini-3.8B from 41.4% to 86.4%, surpassing o1-preview by +4.5% and +0.9%
Source : arxiv.org/html/2501.04...
High-level overview:
The clock generator takes in the PHI0 clock, and splits it four ways, into PHI1-4, which do not overlap.
On the rising edge of PHI1, PHI2, and PHI3, the program counter increments and puts the address on the address bus.
The clock generator takes in the PHI0 clock, and splits it four ways, into PHI1-4, which do not overlap.
On the rising edge of PHI1, PHI2, and PHI3, the program counter increments and puts the address on the address bus.
December 2, 2025 at 2:13 PM
High-level overview:
The clock generator takes in the PHI0 clock, and splits it four ways, into PHI1-4, which do not overlap.
On the rising edge of PHI1, PHI2, and PHI3, the program counter increments and puts the address on the address bus.
The clock generator takes in the PHI0 clock, and splits it four ways, into PHI1-4, which do not overlap.
On the rising edge of PHI1, PHI2, and PHI3, the program counter increments and puts the address on the address bus.
Phi3 architecture is really good
February 13, 2026 at 2:32 AM
Phi3 architecture is really good
interesting news from Docugami, who has always told me they can get LLM-level results from small models by being very focused on their domain (business documents); they're using Phi3-mini and Ministral-8B for mathematical reasoning (which documents like contracts need)
Docugami to Showcase Small Agentic Reasoning Models that Outperform Chat GPT-grade LLMs at Prestigious ICDAR Conference
Docugami Science Team research on Agentic Reasoning; Docugami's small agentic models outperform much larger LLMs, including ChatGPT and GPT-4.
www.docugami.com
September 16, 2025 at 5:55 PM
interesting news from Docugami, who has always told me they can get LLM-level results from small models by being very focused on their domain (business documents); they're using Phi3-mini and Ministral-8B for mathematical reasoning (which documents like contracts need)
Hey #PowerShell people and #AI #PromptEngineers
I made a wrapper for #ollama
One cool feature:
Run language models by name
For example:
phi3 "What is your purpose?"
Have fun!
github.com/StartAutomat...
I made a wrapper for #ollama
One cool feature:
Run language models by name
For example:
phi3 "What is your purpose?"
Have fun!
github.com/StartAutomat...
GitHub - StartAutomating/ollama-powershell: Ollama PowerShell: Play with AI in PowerShell
Ollama PowerShell: Play with AI in PowerShell. Contribute to StartAutomating/ollama-powershell development by creating an account on GitHub.
github.com
April 14, 2025 at 11:40 PM
Hey #PowerShell people and #AI #PromptEngineers
I made a wrapper for #ollama
One cool feature:
Run language models by name
For example:
phi3 "What is your purpose?"
Have fun!
github.com/StartAutomat...
I made a wrapper for #ollama
One cool feature:
Run language models by name
For example:
phi3 "What is your purpose?"
Have fun!
github.com/StartAutomat...
testing #Phi3, #TinyLlama and #Llama3 models on #raspberrypi using #ollama pimylifeup.com/raspberry-pi...
#LLM #SLM #AI #GenAI #Iot #EdgeAI
#LLM #SLM #AI #GenAI #Iot #EdgeAI
How to run Ollama on the Raspberry Pi
Easily run large language models on the Pi
pimylifeup.com
April 26, 2024 at 2:20 PM
testing #Phi3, #TinyLlama and #Llama3 models on #raspberrypi using #ollama pimylifeup.com/raspberry-pi...
#LLM #SLM #AI #GenAI #Iot #EdgeAI
#LLM #SLM #AI #GenAI #Iot #EdgeAI
The instruction register latches a byte each time the counter increments, on the falling edge of PHI1 (instruction byte), PHI2, and PHI3 (operand word).
Thus, the instruction word is 24 bits long.
Thus, the instruction word is 24 bits long.
December 1, 2025 at 11:58 PM
The instruction register latches a byte each time the counter increments, on the falling edge of PHI1 (instruction byte), PHI2, and PHI3 (operand word).
Thus, the instruction word is 24 bits long.
Thus, the instruction word is 24 bits long.
Miの旅/
この施設は学校の一部と見なされ、静かな庭園と子供向けの遊び場があります。夜景を含む景色や日中に訪れるための遊び所も見えています。ですね。
(Powered Google map apis,phi3-3.8mi16,animagine-xl-3.1,LLaVA,etc.basePhoto:Hideyuki Matsubara)
この施設は学校の一部と見なされ、静かな庭園と子供向けの遊び場があります。夜景を含む景色や日中に訪れるための遊び所も見えています。ですね。
(Powered Google map apis,phi3-3.8mi16,animagine-xl-3.1,LLaVA,etc.basePhoto:Hideyuki Matsubara)
September 10, 2024 at 5:57 AM
Miの旅/
この施設は学校の一部と見なされ、静かな庭園と子供向けの遊び場があります。夜景を含む景色や日中に訪れるための遊び所も見えています。ですね。
(Powered Google map apis,phi3-3.8mi16,animagine-xl-3.1,LLaVA,etc.basePhoto:Hideyuki Matsubara)
この施設は学校の一部と見なされ、静かな庭園と子供向けの遊び場があります。夜景を含む景色や日中に訪れるための遊び所も見えています。ですね。
(Powered Google map apis,phi3-3.8mi16,animagine-xl-3.1,LLaVA,etc.basePhoto:Hideyuki Matsubara)
I just ran an AI model #phi3:mini on my home TrunNAS server.
October 27, 2025 at 12:31 AM
I just ran an AI model #phi3:mini on my home TrunNAS server.
Azure AI Studio
aka.ms
December 13, 2024 at 5:24 AM
I just downloaded a larger model (phi3:14b) and this time, it actually produced a PowerShell one-liner instead of a series of PowerShell commands:
November 27, 2024 at 3:47 PM
I just downloaded a larger model (phi3:14b) and this time, it actually produced a PowerShell one-liner instead of a series of PowerShell commands:
Finally got around to make a setup to test the different Ollama models for the things that we use in Dev Proxy. So far, looking at phi3, phi3.5, qwen2.5:3b and llama3.2
January 11, 2025 at 9:10 AM
Finally got around to make a setup to test the different Ollama models for the things that we use in Dev Proxy. So far, looking at phi3, phi3.5, qwen2.5:3b and llama3.2
I began writing this paper, and weeks in, models like Gemma2, Phi3, and Llama3 were released, prompting us to repeat our experiments. By the time the paper was submitted and reviewed, newer models had emerged, and within a year, the cutting edge was already outdated.
December 9, 2024 at 7:26 PM
I began writing this paper, and weeks in, models like Gemma2, Phi3, and Llama3 were released, prompting us to repeat our experiments. By the time the paper was submitted and reviewed, newer models had emerged, and within a year, the cutting edge was already outdated.
Mu's Trip
(reply Powered phi3-3.8mi16,animagine-xl-3.1,etc.)
(reply Powered phi3-3.8mi16,animagine-xl-3.1,etc.)
December 27, 2024 at 4:12 AM
Mu's Trip
(reply Powered phi3-3.8mi16,animagine-xl-3.1,etc.)
(reply Powered phi3-3.8mi16,animagine-xl-3.1,etc.)