mohsen1

3 ms·

1.
by mohsen1 5d ago
> The SAME model may even give different answers to the same prompt when asked multiple times temperature?
2.
by mohsen1 5d ago
see sister comment's response https://news.ycombinator.com/item?id=49786548
3.
by mohsen1 5d ago
This is less true for modern posttrained models. Model identity can be explicitly reinforced during posttraining. Qwen's own finetuning docs include identity training examples, and Qwen models have been trained with system prompts that
4.
by mohsen1 5d ago
I can't find it but saw that if you give Jev English alphabet as choices and ask it in a loop what model it is, it would say Qwen also tried myself: https://console.typesafe.ai/playground?share=shr_1690a3160f1...
5.
by mohsen1 8d ago
There is an open PR for VLLM to do this via DefussionGemma https://github.com/vllm-project/vllm/pull/57250
6.
by mohsen1 8d ago
yup https://github.com/vllm-project/vllm/pull/57250
7.
by mohsen1 8d ago
> A little research before vibe-coding an entire language and compiler could have substantially improved the result because the author would have known what to ask for. A little research before writing and publishing a personal attack li
8.
by mohsen1 11d ago
if you were disappointed by iOS dictation, give it another shot after this update. It has improved a lot. I've been using it a lot more than before
9.
by mohsen1 16d ago
I am speculating but hard to not see that DeepSeek is brewing a full Pro model with those new techniques to come out right around the time of Anthropic and/or OpenAI IPO to tamper the excitement for their offering.
10.
by mohsen1 19d ago
Turkish as well. They moved from Arabic to Latin script.
11.
by mohsen1 19d ago
I absolutely loathe my LG smart TV. Just switching to another input requires 3 button press. I hope someone uses their free time to hack the OS and offer a clean and simple interface to make it a “dumb” TV.
12.
by mohsen1 1mo ago
I have a question about “domain expertise” as a component of future knowledge worker requirements. How does one gain such expertise in a context where thinking is expected to be delegated to AI (shifting from problem solving to question ask
13.
by mohsen1 1mo ago
> The company on Wednesday confirmed speculation that the Ox Alpha model is a new iteration of its GLM series and said it will release the weights for it tonight, in response to queries by Bloomberg News. Seems legit. It's really ha
14.
Issues Reaching Status.claude.com(anthropic.statuspage.io)
3 points by mohsen1 1mo ago|0 comments
15.
by mohsen1 2mo ago
wow i totally missed "It Was Just an Accident". Have to watch asap.
16.
by mohsen1 2mo ago
Hmmm... I had a different experience. They had a fully automated environment where you ha to write test that passes some tests. No human involved. And the time requirement was insanely tight
17.
by mohsen1 2mo ago
That's a lot of words to say you need larger set of questions for today's models. 300 questions won't be enough to find the difference
18.
by mohsen1 2mo ago
> one of the basic tenants of algorithm development was that you can't just brute-force your way towards a solution for some complex problems Mote-Carlo is pretty useful still. Not sure if your statement holds
19.
by mohsen1 2mo ago
Do you think any programmer really understands how their program works end-to-end? At some abstraction layer, we're all clueless. There are many layers between what you type into the text editor and the actual CPU ticks that make your
20.
by mohsen1 2mo ago
RL. Lots of RL
21.
by mohsen1 2mo ago
I thought the same. If your life is so in order that you routinely sleep on the same interval, perhaps your life is not as stressful as others who sleep more chaotically
22.
by mohsen1 3mo ago
This is a Chrome extension that records lots of details in a usage session. Stuff like network calls, console logs, screenshots and also optionally screenshots and user narration Tools like this exist, but every one I tried is uploading the
23.
Show HN: A free and open source session recording Chrome extension
3 points by mohsen1 3mo ago|0 comments
24.
by mohsen1 3mo ago
Thanks! Any pro tips for sharding? I landed on single job because couldn't get cache to work properly for shards to be fast enough to worth it
25.
by mohsen1 3mo ago
I love nextest. without it my CI could take hours https://github.com/tsz-org/tsz/actions/runs/29002057457/job/... watch it running 32.5k unit tests without breaking a sweat!
26.
by mohsen1 3mo ago
This is perfect for my Chrome Extension for recording sessions and capturing screenshots, audio narration and videos. The output is a zip file with everything so if user wants to share they can use this https://github.com/mo
27.
by mohsen1 3mo ago
I have a philosophical problem with adaptive thinking. It’s a dumb guess for how much thinking budget to allocate ahead of thinking. At least in the context of LLMs there is probably no way of knowing how much thinking (token generation) is
28.
by mohsen1 3mo ago
Computer Scientist Explains One Concept in 5 Levels of Difficult https://www.youtube.com/watch?v=fOGdb1CTu5c This video is very beautifully explains it
29.
by mohsen1 3mo ago
This seems to be a new trend. Noticed it with GPT "ultra" in their announcement[1]. I'm with you, a large language model and a system of many language models working together are not the same thing [1] https://news
30.
by mohsen1 3mo ago
It seems like we forget that LLMs are next token prediction systems. Using raw models without instruction following and chat completion bells and whistles will give you a better feeling of what LLMs are. The current interface to LLMs are he

More ›