documentation/autogollark.myco

21 lines
2.4 KiB
Plaintext
Raw Normal View History

2024-10-10 09:11:12 +00:00
Autogollark is an [[emulation]] or primitive [[beta upload]] of [[gollark]] using a proprietary dataset of dumped [[Discord]] messages, [[semantic search]] and [[in-context learning]] on a [[base model]]. Currently, the system uses LLaMA-3.1-405B base in FP8 via Hyperbolic, [[AutoBotRobot]] as a frontend and a custom [[PGVector]]-based search API. While not consistently coherent, Autogollark is able to approximately match personality and typing style.
2024-09-14 06:17:41 +00:00
2024-10-10 09:11:12 +00:00
== TODO
2024-09-14 06:17:41 +00:00
* reformat dataset to include longer-form conversation chunks for increased long-term coherence
* fix emoji/ping formatting
2024-10-10 09:11:12 +00:00
* writeable memory?
== Emergent capabilities
2024-10-10 09:28:06 +00:00
Autogollark has emergently acquired some abilities which were not intended in the design.
2024-10-10 09:28:22 +00:00
* Petulant nonresponse - due to ratelimits in the LLM API, Autogollark will under some circumstances not respond to messages, with error messages being consumed and not exposed to [[users]]. This can be interpreted by [[credulous]] users as choosing not to respond, though this is not believed to be possible (other than cases like responding with `.`, which has not been observed).
2024-10-10 09:28:06 +00:00
* Memorizing links: Autogollark directly experiences past message chunks in context, giving it perfect recall of a small amount of memory. This has memorably included [[Autogollark/Closed Individualism Incident|YouTube videos]] repeated with no context.
* {Limited self-improvement attempts: when told about this architecture, Autogollark will often complain about various limitations and propose vague ideas for improvements.
* Also, Autogollark has previously claimed to be working on LLM chatbots.}
2024-10-10 09:35:18 +00:00
* {Inconsistent inference of own status as a language model chatbot, possibly based on seeing the name "autogollark". Often, Autogollark assumes use of GPT-3.
2024-10-10 09:28:06 +00:00
* Autogollark will also sometimes alternately claim to be the "original" gollark, particularly when interacting with gollark.}
* "Self-reset" from attractor states (e.g. the [[As An AI Language Model Trained By OpenAI]] basin, all caps, etc) after some time passes, because of messages having `HH:MM` timestamps.
2024-10-10 09:31:24 +00:00
* For somewhat [[Waluigi Effect]]-related reasons (past context is strong evidence of capability but weak evidence of incapability), Autogollark has some knowledge [[gollark]] does not, and can speak in a much wider range of languages.
* [[Autogollark/Immortality|Immortality]] via substrate-independence.