Commit Graph

Select branches

Hide Pull Requests

master

Mono Color

• 0093a70c51 slight rewrite, mostly as functional as before master mrq 2023-05-22 15:40:11 +0000
• 6fa2c18fb1 reworked saving/loading agent by just saving the memory documents themselves and adding them on load, rather than serializing the entire memory object and it breaking between systems / wizard prompt tune / more tunings mrq 2023-05-15 07:28:04 +0000
• 287406e7ba tunings mrq 2023-05-09 23:57:54 +0000
• f13d05dbb2 more tuning mrq 2023-05-03 01:46:55 +0000
• 8eaecaf643 updating for new langchain, more tunes mrq 2023-05-03 01:01:58 +0000
• e152cd98a4 updated requirements because I had installed this in WSL2 mrq 2023-05-03 00:26:37 +0000
• 41e48497cd more tunes mrq 2023-05-01 05:47:35 +0000
• f10ea1ec2a added prompt tuning for superCOT (which 33B seems to be the best approach for a local LLM) mrq 2023-04-30 22:56:02 +0000
• 089b7043b9 more tuning mrq 2023-04-30 20:00:46 +0000
• e9abd9e73f Swapped to a much simpler way of formatting prompts given a finetune by recording prompts as system/user/assistant dicts, then combine them according to provided finetune mrq 2023-04-30 17:57:53 +0000
• 0964f48fc0 ui fixes mrq 2023-04-30 03:30:46 +0000
• f9cfd1667f I think I better tuned the prompts for vicuna, I forget (and added the licenses for Langchain code because I can't be assed to inject a bunch of functions now) mrq 2023-04-30 01:27:34 +0000
• a1cb43da5e speedups by terminating early and not having short observations take forever from ramblings mrq 2023-04-29 22:04:18 +0000
• 9e0fd8d79c more changes to make LLaMAs cooperate better with better tuned prompts, just in time for todd to be put down mrq 2023-04-29 18:48:33 +0000
• b553ffbc5f added rudimentary web UI mrq 2023-04-29 05:54:55 +0000
• b35f94d319 an amazing commit mrq 2023-04-29 04:14:56 +0000
• 10cf7746d8 Initial commit mrq 2023-04-29 03:37:11 +0000