zlacker

[return to "Mistral 7B Fine-Tune Optimized"]
◧
1. m3kw9+zH[view] [source] 2023-12-21 00:17:10
>>tosh+(OP)
I’m really struggling to find a use case for these local models when even ChatGPT 3.5 can do it as good as any of them so far.
◧◩
2. coder5+KK[view] [source] 2023-12-21 00:40:06
>>m3kw9+zH
The article shows (fine tuned) Mistral 7B outperforming GPT-4, never mind GPT-3.5.
◧◩◪
3. m3kw9+kR[view] [source] 2023-12-21 01:43:12
>>coder5+KK
This model is not close to even 3.5 from when I used it. It first of all does not follow instructions properly and it just runs on and on
◧◩◪◨
4. coder5+6T[view] [source] 2023-12-21 02:05:25
>>m3kw9+kR
What you're describing is the behavior you get from any base model that has not been instruction-tuned. The article is clear that this model is not for "direct use". It needs tuning for a specific application.
◧◩◪◨⬒
5. m3kw9+c01[view] [source] 2023-12-21 03:24:46
>>coder5+6T
how does one fine tune it to follow instructions? I would have thought they have open source training set for these instruction-follow fine tunes?
[go to top]