Rietenpetardat: Difference between revisions

1,520 bytes added ,  10 months ago
added commands & the return
(added commands & the return)
Line 15: Line 15:
Its profile picture was generated with Stable Diffusion by [[Razvan5576]].
Its profile picture was generated with Stable Diffusion by [[Razvan5576]].


== Development ==
== Commands ==
The following refers to the latest version of the rietenpetardat as of 11th of August 2023.
 
* /gptconfig - Prints the current configuration of the GPT model
* /identitate <identitate> - Sets the identity to the paramater (only for moderators)
* /parametru - Sets a paramater from the list below (only for moderators)
** modgen - Selects the way which a message is generated
*** Organic - Text is generated until a response from the identity is generated
*** Forțat - A response belonging to the identity is appended to the next line, leaving the AI to forcefully generate a message from that identity
** temperatura
*** A floating point number indicating the temperature (randomness) of the generation
** top_k
*** The first top_k tokens, sorted in order of appearances
** seed
*** An integer number used for generating random numbers for the generation
* /reinitializare - Reinitialises the model. Necessary when changing some options (only for moderators)
* /resetmemorie - Clears the memory of past messages (only for moderators)
 
== Development history ==


=== Initial version ===
=== Initial version ===
Line 43: Line 61:


=== Temporary shutdown ===
=== Temporary shutdown ===
As of April 24, 2023 the GPT-3 chat functions and the ChatGPT command no longer work because Edward's payment method started rejecting payments from OpenAI, making the API inaccessible.
On 24th of April, 2023 the GPT-3 chat functions and the ChatGPT command no longer work because Edward's payment method started rejecting payments from OpenAI, making the API inaccessible.
 
=== The return ===
Since the shutdown, Edward has been searching for a way to train an AI which can be fully self-hosted. An attempt has been made to train a language model using [https://marian-nmt.github.io/ marian-nmt], but this has not been successful. On 18th of July he began experimenting with [https://github.com/karpathy/nanoGPT nanoGPT] and he was able to train a model from scratch on data from past chats from the OKPR Discord server. The next day, a bare-bones but operational version was finished and released for a few hours on the OKPR Discord server for testing.  


=== Future plans ===
The Discord bot is now developed in Python using the [https://pycord.dev/ Pycord] library. Tests of the bot are done mainly on a private server.
Edward is planning on training a custom model as described below. An attempt has been made to train a language model using [https://marian-nmt.github.io/ marian-nmt], but this has not been successful. He is now planning to train a GPT-style model.


== Technical details and research ==
== Technical details and research ==
115

edits