I have been developing a Tinder swiping and chatting bot trough their iOS mobile API, it all works well, but some accounts get the arkose captcha challenge when I try to refresh the Auth Token. This is the response I get from the API call that should refresh the auth token:
{'error': {'code'...
Yo bro to me what worked before stopped working right after the most recent update, than it started working again in the past few days
Hope it has any meaning to u / helps
yes so I was able to run my model using simple ollama and hosting on a cheap Hetzner server for like 45$ / month
I am using 7B from Mistral as a base model
hey thanks for those good advices. I managed to gather some data than I used an AI model to replicate same style of messages in order to expend on that small dataset I had. managed to get a total of 500 rows (which is not optimal nor a lot a lot. would need way more to actually make it...
Thanks for sharing your experience. This stull can be pretty overwhelming when there are millions of things u can do and not even sure what is the best approach for your needs.
I am building somewhat of a chatbot so I can really predict the flow of the queries, but I will try to keep it from...
Thanks bro this was exactly wat I was looking for. I failed first try because the training file I used was formatted something like this:
user input: *blablabla*
reply: *some stuff*
so like I gave it hypothetical inputs and the replies I would want it to give to them. this way I thought it...
Hi, I am trying something quite similar, I wish to create a chatbot that does not act like the common "Helpful Assistant", but rather impersonate a real human, for example a salesperson. For that I need to make the AI break character and switch to a different way of "speaking" to say.
My guess...
As I feel like this is a tangent to the topic of this discussion, anyone has knowledge on how one can tarin an existing LLM in order to make it "sound" in a certain way? For example if I want to make it chat like an actor per say, and not like a "Helpful Assistant". What should be done to...
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.