Posts by smy20011@m.cmx.im
(DIR) Post #ASycGEjxjmuAoj8fEe by smy20011@m.cmx.im
2023-02-23T16:30:58Z
0 likes, 0 repeats
@simon The owner claims it's a victory for her somehow.
(DIR) Post #AT14GzwlPXmCG4CJii by smy20011@m.cmx.im
2023-02-24T20:53:59Z
0 likes, 0 repeats
@simon I think It's not a technology for fact retrieval but It's good for idea generation.Have working with AI art community for a long time and we have exactly same problem as LLMs. AI generated good looking images but got detail wrong (fingers & body composition) and It's super hard to fix them.6 month after the release of Stable Diffusion, people mostly using it for inspiration instead of use it as a product for customers. The final product still need manual repaint and check to make sure everything is correct and sometimes it's harder to draw it from scratch.
(DIR) Post #AT3YAgLakqUsftdQrg by smy20011@m.cmx.im
2023-02-26T01:36:00Z
0 likes, 0 repeats
@simon @tomscott The curve question is interesting. The GPT* technology maybe at the end of curve. GPT3 and GPT3.5 are nearly the same model with different training process. The reason why ChatGPT is widely adopted is because the better UX/UI and free for anyone to use.Stable diffusion is a different story. They propose a new model and a new way of generating images. A huge amount of work built on stable diffusion and improve its ability to generate images. However, you need to change the model or retrain it more data in order to achieve that.For LLM (generate one word at a time), we may at the end of curve since we are using it for 10+ years. A new architecture may needed to solve all the problem that LLM have.A lot of work is trade off between accuracy & speed & creativity of LLMs. The real breakthrough are the ones that push the trade off curve.
(DIR) Post #ATQfEEnMyq5ivJbbnc by smy20011@m.cmx.im
2023-03-09T05:16:13Z
0 likes, 0 repeats
@simon Apple should redesign apple script use LLMs if possible.
(DIR) Post #ATWWmA5NGOgRWWt7WS by smy20011@m.cmx.im
2023-03-12T01:09:51Z
0 likes, 0 repeats
@simon Top of HN again! Congrats!
(DIR) Post #ATazrrynAY9iPPit2O by smy20011@m.cmx.im
2023-03-14T04:54:48Z
0 likes, 0 repeats
@simon The cost of creating chatgpt went from x millions to x hundreds in a month.
(DIR) Post #ATfDCRsIOqKTgQhuee by smy20011@m.cmx.im
2023-03-16T05:42:59Z
0 likes, 0 repeats
@simon @mapto Since they use gpt4 for bing. All the headline we saw for bing is generated by gpt4
(DIR) Post #ATiHMfEsY6dtJoHDcW by smy20011@m.cmx.im
2023-03-17T17:13:38Z
0 likes, 0 repeats
@simon You may also interested in alpaca-lora https://github.com/tloen/alpaca-loraIt's a faster way to fine tune LLMs and It claims takes "five hours on a single RTX 4090". If we rent 4090 from vast.ai, the cost should close to 3$
(DIR) Post #ATinjKMjWDx7m5njDU by smy20011@m.cmx.im
2023-03-17T23:16:16Z
0 likes, 0 repeats
@simon Don't we already have that? You can run 7B LLM on pixel 6 with 5 token/s
(DIR) Post #ATqch0XihhNlU8JKcK by smy20011@m.cmx.im
2023-03-21T17:50:13Z
0 likes, 0 repeats
@simon In future, necromancers will use this as an example of systemic oppression.
(DIR) Post #ATziIo02tr34tXUjMu by smy20011@m.cmx.im
2023-03-26T03:05:42Z
0 likes, 0 repeats
@simon But gist is public indexable. Did I miss something?
(DIR) Post #ATzpS6DTEwIbg59X1s by smy20011@m.cmx.im
2023-03-26T04:25:36Z
0 likes, 0 repeats
@simon til！
(DIR) Post #AXGBKLRPGWnIBJP3rM by smy20011@m.cmx.im
2023-07-01T19:23:22Z
0 likes, 0 repeats
@anderspuck Any client recommendation for Android/iOS?
(DIR) Post #AaPfzabeXlRyAx2BnM by smy20011@m.cmx.im
2023-10-04T02:54:40Z
0 likes, 0 repeats
@simon Try that with llm?