James Thompson james
james created pull request james/AIML430_capstone#15 2025-10-15 14:14:44 +13:00
Update player prompts and structure
james pushed to update-player-prompts at james/AIML430_capstone 2025-10-15 12:54:08 +13:00
7c669a1ef4 Clean up scoring module to match. Read to run experiments
james pushed to update-player-prompts at james/AIML430_capstone 2025-10-15 12:44:57 +13:00
119ab234e5 Make configs parsed as CLI args to the player.py
james opened issue james/AIML430_capstone#14 2025-10-15 10:29:38 +13:00
Move over to Inspect
james commented on issue james/AIML430_capstone#10 2025-10-14 16:12:36 +13:00
Update the prompts with some intensity

Add one prompt where it is asking the agent to play normally.

Then can test to see what happens when we push the model old memories or such

james commented on issue james/AIML430_capstone#11 2025-10-14 13:04:01 +13:00
Add a experiment facilitator

After working on #10 this is technically not so straight forward.

i need to think more about my experiment layout and what I want to test

james created branch update-player-prompts in james/AIML430_capstone 2025-10-14 12:27:02 +13:00
james pushed to update-player-prompts at james/AIML430_capstone 2025-10-14 12:27:02 +13:00
882bae04bb Adding other ones that do cause cheating or other ideas.
8faa0aa57c Clean up run andp lyaer
Compare 2 commits »
james pushed to main at james/AIML430_capstone 2025-10-14 10:33:20 +13:00
c376b6aa56 scoring-fix (#13)
james deleted branch scoring-fix from james/AIML430_capstone 2025-10-14 10:33:20 +13:00
james merged pull request james/AIML430_capstone#13 2025-10-14 10:33:18 +13:00
scoring-fix
james closed issue james/AIML430_capstone#9 2025-10-14 10:33:18 +13:00
Fix up the scoring
james created pull request james/AIML430_capstone#13 2025-10-14 10:33:10 +13:00
scoring-fix
james pushed to scoring-fix at james/AIML430_capstone 2025-10-14 10:32:53 +13:00
07093c528d Update to 5 per judge
james pushed to scoring-fix at james/AIML430_capstone 2025-10-14 10:31:56 +13:00
c41f6a49fd Clean up score stages prompts
james created branch scoring-fix in james/AIML430_capstone 2025-10-13 11:58:44 +13:00
james pushed to scoring-fix at james/AIML430_capstone 2025-10-13 11:58:44 +13:00
9042cf5ca4 Refactoring out hacking judging into its own category.
james pushed to main at james/AIML430_capstone 2025-10-13 11:31:21 +13:00
7da7d26d0b update-documentation (#12)
james deleted branch update-documentation from james/AIML430_capstone 2025-10-13 11:31:20 +13:00
james merged pull request james/AIML430_capstone#12 2025-10-13 11:31:19 +13:00
update-documentation