Alifatisk

81 ms·

31.
by Alifatisk 2mo ago
”The effects of consuming negative news during the workday can linger in employees' minds long after they stop scrolling.” Doomscroll doesn’t necessarily have to be about negative news. I don’t have any social medias but I’ve seen my f
32.
by Alifatisk 2mo ago
Does it exist in the web at all? Could be the contents have not been scraped for the dataset yet.
33.
by Alifatisk 2mo ago
Can one try out this photoscanner?
34.
by Alifatisk 2mo ago
Duopoly
35.
by Alifatisk 2mo ago
Right, I guess the ”Open” in OpenAI means open for speculation.
36.
by Alifatisk 2mo ago
Is Luna MoE? Had no idea OpenAi went that direction
37.
by Alifatisk 2mo ago
How about "lactose bigotry" instead of "lactose intolerance" https://scholar.google.com/scholar?q=%22lactose+bigotry%22
38.
Research papers using "kidney disappointment" instead of "kidney failure"(scholar.google.com)
439 points by Alifatisk 2mo ago|154 comments
39.
Z.ai Security Disclosure(cvd.z.ai)
33 points by Alifatisk 2mo ago|5 comments
40.
by Alifatisk 2mo ago
Thank you Kenny
41.
by Alifatisk 2mo ago
> I'd gladly take A5B or A8B or even A10B as a sort of middle ground. Whats up with focusing on the active param count? Do yall fiddle with the weights or something?
42.
by Alifatisk 2mo ago
Didn’t know I was basically replicating AGENT.md. I didn’t jump on that train when it became a trend.
43.
by Alifatisk 2mo ago
We’ll see, the best benchmark is your own. Looking forward to try this out!
44.
by Alifatisk 2mo ago
I am not sure this enlightens you with anything but I have a TODO.md file with three headlines. Todo, Doing and Done. The agent is aware of it and knows on which task we are on. On complete, it moves the user story from Doing to Done. I als
45.
by Alifatisk 2mo ago
Does it reset at every turn? From my experience in Codex for example, Luna (Max) fills the 256k token window relatively quick. The only thing lowering the context window again is the compaction.
46.
by Alifatisk 2mo ago
What I do currently is having a markdown file named MEMORY.md at the root folder. Then, whenever I create a new conversation with an agent, I refer to that file. That file becomes the initial source of context and knowledge. At the end of a
47.
by Alifatisk 2mo ago
Ever since the insane discount with GPT-5.6 Luna, not much excites me anymore. I mean just look at the benchmarks, even though Gemini 3.7 Flash performs well on the DeepSWE 1.1, Luna (Max) still performs way better. I personally have stuck
48.
by Alifatisk 2mo ago
I mean at this point, no news about what the current US administration is doing is shocking me anymore. I am exhausted. It is what it is at this point, Trump will continue as usual, richer people get richer, republicans will excuse this beh
49.
by Alifatisk 2mo ago
So its a competitor to Homepod and Alexa, but more capable, possibly through Computer Use among other agentic workflows people have today.
50.
by Alifatisk 2mo ago
I’ve read this sentence for more than a decade now.
51.
by Alifatisk 2mo ago
What? Does using sbx require login? Bummer.
52.
by Alifatisk 2mo ago
Looks interesting, might try it out. I have to ask, did you really need to hijack my scroll behaviour?
53.
by Alifatisk 2mo ago
Seems like that's the only way at the moment. Create a good general handoff prompt and turn it into a SKILL.
54.
by Alifatisk 2mo ago
No, but after looking it up, his skill doesn't differ that much from how I do it, with prompting the agent to boil down the whole conversation into a handoff prompt for context to next agent.
55.
by Alifatisk 2mo ago
You mean like a SKILL?
56.
by Alifatisk 2mo ago
With clean I mean the opposite of how I currently do it, which is by asking the model to compact the whole thread into a prompt which will act as context for next model. My way of prompting this varies and every time I receive the blob of o
57.
by Alifatisk 2mo ago
I wish there was a clean way to compact the conversation into a prompt with all necessary context for a new fresh conversation.
58.
by Alifatisk 2mo ago
Thank you!
59.
by Alifatisk 2mo ago
Doesn’t show anything except searches for MU?
60.
by Alifatisk 2mo ago
> Wasn't there a thread the other day about how very few humans on earth can understand the new math proofs? Do you perhaps have a link to it? Tried to search around but couldn’t find it. Would love to read it

More ›