Commit graph

363 commits

Author SHA1 Message Date
AlxAI
357aeea48e v2 and v3 of holds 2025-07-17 20:11:24 -04:00
AlxAI
eeeb312d40 hold reduction 2025-07-17 14:22:07 -04:00
sam-paech
d9b641b604 optimised prompts v1.0 2025-07-17 22:50:43 +10:00
sam-paech
d785ada9c0 use standard function for generating board state str 2025-07-17 22:02:12 +10:00
sam-paech
77086f5c77 display unit & sc counts 2025-07-17 14:38:51 +10:00
sam-paech
5bb008bc34 default to include possible moves summary in order generation 2025-07-17 14:24:59 +10:00
sam-paech
7c59153ab4 fix missing data in gamehistory object on resume 2025-07-16 16:51:48 +10:00
sam-paech
4a19548cac add active playstyle tips to default system prompts 2025-07-15 14:44:44 +10:00
sam-paech
7735831513 fix order classification bug 2025-07-14 22:30:41 +10:00
sam-paech
86563587a5 proper fix for game score 2025-07-14 19:55:02 +10:00
sam-paech
fb05356fe7 fix game score calculation 2025-07-14 11:55:40 +10:00
sam-paech
62521b1e55 add condensed possible move summary for ablations 2025-07-13 12:12:36 +10:00
sam-paech
b4a56126ec state update fixes & streamline prompts 2025-07-12 10:17:17 +10:00
sam-paech
1f154a7073 fixes for state updates 2025-07-10 21:52:22 +10:00
sam-paech
3d591a11d1 critical state & plotting fixes 2025-07-10 12:20:30 +10:00
sam-paech
70a876bcee add relationship plots 2025-07-10 10:35:17 +10:00
sam-paech
af3fb8ce48 fix sentiment calcs 2025-07-10 07:49:47 +10:00
sam-paech
2338e3c6e6 bug fixes 2025-07-10 07:06:34 +10:00
Tyler Marques
754cf8e23b
Merge pull request #52 from EveryInc/experiment-runner-updates
Harness defaults, client.py updates & new experiment runner stats
2025-07-04 10:20:24 -07:00
Tyler Marques
ca5e415d89
Removing file accidently added
Signed-off-by: Tyler Marques <me@tylermarques.com>
2025-07-04 09:20:57 -07:00
sam-paech
e351aa3841 savegame fix + chart updates 2025-07-05 00:27:02 +10:00
sam-paech
7edc7c465f fix prompt errors & add per-power prompt dir functionality 2025-07-04 11:31:57 +10:00
sam-paech
22cf3202c7 cleanup 2025-07-04 09:35:57 +10:00
sam-paech
05016cafba clients raise errors to run_llm_and_log 2025-07-04 09:26:46 +10:00
sam-paech
1b16579c54 fix imports 2025-07-04 08:59:42 +10:00
sam-paech
73a3c0596a resolving issues raised in pr 2025-07-04 08:38:57 +10:00
Tyler Marques
ad817695ef
Ruff format and check 2025-07-03 14:27:50 -07:00
Tyler Marques
aadedd6512
Moving more environment variables to central config 2025-07-03 12:50:49 -07:00
Tyler Marques
540c2003e8
Adding PowerEnum as a seperate model
The PowerEnum correctly handles some misspellings. It can be easily
expanded to handle more within the _POWER_ALIASES dict.
2025-07-03 12:26:50 -07:00
sam-paech
a241e34496 fix to respect model ids passed as args when resuming 2025-07-03 09:59:47 +10:00
sam-paech
4fc1f370be allow specification of base url + api key for each model 2025-07-03 08:55:10 +10:00
sam-paech
f29ac9c1c5 all client types can be specified with a prefix in front of the model id 2025-07-03 07:41:29 +10:00
sam-paech
1cb24f1884 set minimalist prompts as harness defaults 2025-07-03 07:27:48 +10:00
sam-paech
cbd475c0c2 add new comparative stats to experiment runner 2025-07-03 03:30:47 +10:00
Tyler Marques
7905f9a0b8
Merge pull request #51 from peregrinates/ordersdata 2025-07-01 20:12:45 -07:00
peregrinates
00ce085e35
make a long form dataframe of each order in game from log data 2025-07-01 22:49:00 -04:00
AlxAI
46cabc1786 circular import 2025-06-30 12:12:52 +02:00
AlxAI
01a0f227a9 Update experiment_runner.py 2025-06-30 12:08:56 +02:00
AlxAI
7b633a0ec8 making phase summaries and unformatted prompts params in lm_game as well as making the models changeable in utils 2025-06-30 12:03:37 +02:00
AlxAI
861a5a222f fixed un-needed changes 2025-06-30 09:32:26 +02:00
AlxAI
0f01618354 Update formatter.py 2025-06-30 08:01:35 +02:00
AlxAI
6e4821735d first attempt at two part orders, unformatted -> formatted 2025-06-29 23:05:48 +02:00
Tyler Marques
01c6dde114
Merge pull request #45 from EveryInc/order-history
Order history
2025-06-29 10:10:37 -07:00
Tyler Marques
dd7aba87b0
Merge pull request #44 from EveryInc/simple_prompts
add simplified prompts
2025-06-29 10:07:00 -07:00
Tyler Marques
43a2482d44
Merge pull request #48 from EveryInc/feature/default-reviewers
Adding code owners for default PR reviewers
2025-06-29 10:03:42 -07:00
Tyler Marques
282598d173
Adding code owners for default PR reviewers
Signed-off-by: Tyler Marques <me@tylermarques.com>
2025-06-29 10:00:43 -07:00
sam-paech
b5a84867a1 add order history 2025-06-29 01:53:03 +10:00
sam-paech
ebf26cf8a6 add simplified prompts 2025-06-27 14:42:05 +10:00
Tyler Marques
0bd909b30b
Adding run with uv instructions
Signed-off-by: Tyler Marques <me@tylermarques.com>
2025-06-23 09:45:23 -07:00
Tyler Marques
46525efe6b
Merge pull request #43 from EveryInc/feature/uv-dependences
Repo cleanup and uv tool
2025-06-23 09:34:50 -07:00