API tools faq
paste
Zero989

AS66 Zero shot Zero pre-training

Zero989
Nov 1st, 2025 (edited)
124
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 67.26 KB | None | 0 0
raw download clone embed print report
  1. ================================================================================
  2. FinalGift Hybrid Scientist Agent - Dual-Process Intelligence
  3.  
  4. INTENT MODES:
  5. LEARN: Exploring new strategies, hypothesis testing
  6. EXPLOIT: Executing known patterns, using learned knowledge
  7.  
  8. EXECUTION STYLES:
  9. INTUITIVE: Fast policy network (reflexive, System 1)
  10. DELIBERATE: Slow world-model planning (analytical, System 2)
  11.  
  12. Execution blend controlled by capability growth (dBI)
  13.  
  14. TRAINING MODE: GENERALIST
  15. Games: 6 (as66-821a4dcad9c2, ft09-b7249182, lp85-e179de251, ls20-e98196ab, sp80-445eab21, vc33-cbded52d)
  16. Episodes: Unlimited
  17. Checkpoint: checkpoint_latest.pt
  18. WEIGHTED SAMPLING: 3x more attempts for unsolved games (<50% win rate)
  19. ================================================================================
  20.  
  21. ================================================================================
  22. SYSTEM-3 REFLECTIVE META-CONTROLLER: ENABLED
  23. Reflective Cycle Frequency: Every 25 episodes
  24. Safe Reconfiguration Actions: 14
  25. - zeno_alpha (mode sensitivity)
  26. - reward_weight (RL vs self-supervised)
  27. - empowerment_weight (curiosity)
  28. - zeno_beta (attractor smoothing)
  29. - learning_rate (adaptation speed)
  30. Rollback Threshold: dBI < -0.01
  31. ================================================================================
  32.  
  33.  
  34. Checking for checkpoint: checkpoint_latest.pt
  35. File exists: False
  36.  
  37. ================================================================================
  38. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 1/15
  39. ================================================================================
  40. [Ep0] EXPLORATION: No memory replay
  41. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  42. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  43. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  44. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  45. [HER] Applied hindsight relabeling to failed episode (reward=0.00, length=49)
  46. [Ep0] NOT STORED: No level progress (reward=0.00)
  47.  
  48. [Running Abstractive Planning...]
  49. [HYBRID RULES] Allocated rule bank for as66-821a4dcad9c2 (1 games total)
  50. Generated 32 abstract experiences (dreams).
  51.  
  52. [Training Models...]
  53.  
  54. [Training from Abstract Experiences...]
  55. Abstract Value Loss: 11.9153
  56. Abstract Policy Loss: 1.7911
  57. [RETRY] Episode ended with no reward. Failed attempt.
  58. [RETRY] Attempt 1/15. Retrying...
  59.  
  60. ================================================================================
  61. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 2/15
  62. ================================================================================
  63. [Ep0] EXPLORATION: No memory replay
  64. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  65. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  66. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  67. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  68. [API REWARD] score: 0 -> 1, reward: 1.00
  69. [LEVEL 1 COMPLETE @ step 14] Episode reward: 1.0
  70. [MEMORY] NEW level 1 beaten in 14 steps
  71. [TRUNCATED] Storing 15/77 actions (up to and including reward step)
  72. [Ep0] STORED MEMORY: score=1.00, length=15, levels=1 (with transfer data)
  73. [SAVED] episodic_memory.pt updated immediately
  74.  
  75. [CRYSTALLIZATION ACTIVE] Game as66-821a4dcad9c2: Reached score 1.0/9.0
  76. [CRYSTALLIZATION] Progress: 11%, strength: 0.11
  77. [CRYSTALLIZATION] Best: 1.0/9.0, Strength=0.11, Dampening=0.91
  78. [SAVED] crystallization.pt updated (new best: 1.0)
  79.  
  80. [Running Abstractive Planning...]
  81. Generated 32 abstract experiences (dreams).
  82.  
  83. [Training Models...]
  84. [dBI DEBUG as66-821a4dcad9c2] Total history: 2, Valid BIs: 2
  85. [dBI DEBUG] Last 5 BIs: ['0.0252', '0.0354']
  86. [dBI DEBUG] dbi_raw = 0.035355 - 0.025163 = 0.010193
  87.  
  88. [Training from Abstract Experiences...]
  89. Abstract Value Loss: 15.8189
  90. Abstract Policy Loss: 1.8064
  91. [RETRY] NEW PROGRESS! Reached 1 NEW levels. Best: 0 -> 1/9
  92. [RETRY] Resetting to Attempt 1/15 due to progress. Retrying...
  93.  
  94. ================================================================================
  95. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 1/15
  96. ================================================================================
  97. [INITIAL SCORE] API started at score=1, initializing levels_completed=1
  98. [Ep0] EXPLORATION: No memory replay
  99. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  100. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  101. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  102. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  103. [Ep0] NOT STORED: No level progress (reward=1.00)
  104. [CRYSTALLIZATION] Best: 1.0/9.0, Strength=0.11, Dampening=0.91
  105.  
  106. [Running Abstractive Planning...]
  107. Generated 32 abstract experiences (dreams).
  108.  
  109. [Training Models...]
  110. [dBI DEBUG as66-821a4dcad9c2] Total history: 3, Valid BIs: 3
  111. [dBI DEBUG] Last 5 BIs: ['0.0252', '0.0354', '0.0828']
  112. [dBI DEBUG] dbi_raw = 0.082848 - 0.035355 = 0.047493
  113.  
  114. [Training from Abstract Experiences...]
  115. [RETRY] No new levels beaten. Best remains: 1/9
  116. [RETRY] Attempt 1/15. Retrying...
  117.  
  118. ================================================================================
  119. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 2/15
  120. ================================================================================
  121. [INITIAL SCORE] API started at score=1, initializing levels_completed=1
  122. [Ep0] EXPLORATION: No memory replay
  123. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  124. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  125. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  126. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  127. [Ep0] NOT STORED: No level progress (reward=1.00)
  128. [CRYSTALLIZATION] Best: 1.0/9.0, Strength=0.11, Dampening=0.91
  129.  
  130. [Running Abstractive Planning...]
  131. Generated 32 abstract experiences (dreams).
  132.  
  133. [Training Models...]
  134. [dBI DEBUG as66-821a4dcad9c2] Total history: 4, Valid BIs: 4
  135. [dBI DEBUG] Last 5 BIs: ['0.0252', '0.0354', '0.0828', '0.1452']
  136. [dBI DEBUG] dbi_raw = 0.145243 - 0.082848 = 0.062395
  137.  
  138. [Training from Abstract Experiences...]
  139. [RETRY] No new levels beaten. Best remains: 1/9
  140. [RETRY] Attempt 2/15. Retrying...
  141.  
  142. ================================================================================
  143. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 3/15
  144. ================================================================================
  145. [INITIAL SCORE] API started at score=1, initializing levels_completed=1
  146. [Ep0] EXPLORATION: No memory replay
  147. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  148. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  149. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  150. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  151. [Ep0] NOT STORED: No level progress (reward=1.00)
  152. [CRYSTALLIZATION] Best: 1.0/9.0, Strength=0.11, Dampening=0.91
  153.  
  154. [Running Abstractive Planning...]
  155. Generated 32 abstract experiences (dreams).
  156.  
  157. [Training Models...]
  158. [dBI DEBUG as66-821a4dcad9c2] Total history: 5, Valid BIs: 5
  159. [dBI DEBUG] Last 5 BIs: ['0.0252', '0.0354', '0.0828', '0.1452', '0.1238']
  160. [dBI DEBUG] dbi_raw = 0.123755 - 0.145243 = -0.021488
  161.  
  162. [Training from Abstract Experiences...]
  163. [RETRY] No new levels beaten. Best remains: 1/9
  164. [RETRY] Attempt 3/15. Retrying...
  165.  
  166. ================================================================================
  167. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 4/15
  168. ================================================================================
  169. [INITIAL SCORE] API started at score=1, initializing levels_completed=1
  170. [Ep0] EXPLORATION: No memory replay
  171. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  172. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  173. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  174. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  175. [Ep0] NOT STORED: No level progress (reward=1.00)
  176. [CRYSTALLIZATION] Best: 1.0/9.0, Strength=0.11, Dampening=0.91
  177.  
  178. [Running Abstractive Planning...]
  179. Generated 32 abstract experiences (dreams).
  180.  
  181. [Training Models...]
  182. [dBI DEBUG as66-821a4dcad9c2] Total history: 6, Valid BIs: 6
  183. [dBI DEBUG] Last 5 BIs: ['0.0354', '0.0828', '0.1452', '0.1238', '0.1065']
  184. [dBI DEBUG] dbi_raw = 0.106548 - 0.123755 = -0.017207
  185.  
  186. [Training from Abstract Experiences...]
  187. [RETRY] No new levels beaten. Best remains: 1/9
  188. [RETRY] Attempt 4/15. Retrying...
  189.  
  190. ================================================================================
  191. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 5/15
  192. ================================================================================
  193. [INITIAL SCORE] API started at score=1, initializing levels_completed=1
  194. [Ep0] EXPLORATION: No memory replay
  195. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  196. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  197. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  198. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  199. [Ep0] NOT STORED: No level progress (reward=1.00)
  200. [CRYSTALLIZATION] Best: 1.0/9.0, Strength=0.11, Dampening=0.91
  201.  
  202. [Running Abstractive Planning...]
  203. Generated 32 abstract experiences (dreams).
  204.  
  205. [Training Models...]
  206. [dBI DEBUG as66-821a4dcad9c2] Total history: 7, Valid BIs: 7
  207. [dBI DEBUG] Last 5 BIs: ['0.0828', '0.1452', '0.1238', '0.1065', '0.1090']
  208. [dBI DEBUG] dbi_raw = 0.109029 - 0.106548 = 0.002480
  209.  
  210. [Training from Abstract Experiences...]
  211. [RETRY] No new levels beaten. Best remains: 1/9
  212. [RETRY] Attempt 5/15. Retrying...
  213.  
  214. ================================================================================
  215. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 6/15
  216. ================================================================================
  217. [INITIAL SCORE] API started at score=1, initializing levels_completed=1
  218. [Ep0] EXPLORATION: No memory replay
  219. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  220. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  221. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  222. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  223. [API REWARD] score: 1 -> 2, reward: 1.00
  224. [LEVEL 2 COMPLETE @ step 10] Episode reward: 2.0
  225. [CRYSTALLIZATION] Game as66-821a4dcad9c2: NEW BEST 2.0/9.0 (strength: 0.22)
  226. [CRYSTALLIZATION] Best: 2.0/9.0, Strength=0.22, Dampening=0.82
  227. [SAVED] crystallization.pt updated (new best: 2.0)
  228.  
  229. [Running Abstractive Planning...]
  230. Generated 32 abstract experiences (dreams).
  231.  
  232. [Training Models...]
  233. [dBI DEBUG as66-821a4dcad9c2] Total history: 8, Valid BIs: 8
  234. [dBI DEBUG] Last 5 BIs: ['0.1452', '0.1238', '0.1065', '0.1090', '0.0846']
  235. [dBI DEBUG] dbi_raw = 0.084611 - 0.109029 = -0.024417
  236.  
  237. [Training from Abstract Experiences...]
  238. [RETRY] NEW PROGRESS! Reached 1 NEW levels. Best: 1 -> 2/9
  239. [RETRY] Resetting to Attempt 1/15 due to progress. Retrying...
  240.  
  241. ================================================================================
  242. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 1/15
  243. ================================================================================
  244. [INITIAL SCORE] API started at score=2, initializing levels_completed=2
  245. [Ep0] EXPLORATION: No memory replay
  246. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  247. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  248. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  249. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  250. [Ep0] NOT STORED: No level progress (reward=2.00)
  251. [CRYSTALLIZATION] Best: 2.0/9.0, Strength=0.22, Dampening=0.82
  252.  
  253. [Running Abstractive Planning...]
  254. Generated 32 abstract experiences (dreams).
  255.  
  256. [Training Models...]
  257. [dBI DEBUG as66-821a4dcad9c2] Total history: 9, Valid BIs: 9
  258. [dBI DEBUG] Last 5 BIs: ['0.1238', '0.1065', '0.1090', '0.0846', '0.1647']
  259. [dBI DEBUG] dbi_raw = 0.164687 - 0.084611 = 0.080075
  260.  
  261. [Training from Abstract Experiences...]
  262. [RETRY] No new levels beaten. Best remains: 2/9
  263. [RETRY] Attempt 1/15. Retrying...
  264.  
  265. ================================================================================
  266. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 2/15
  267. ================================================================================
  268. [INITIAL SCORE] API started at score=2, initializing levels_completed=2
  269. [Ep0] EXPLORATION: No memory replay
  270. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  271. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  272. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  273. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  274. [Ep0] NOT STORED: No level progress (reward=2.00)
  275. [CRYSTALLIZATION] Best: 2.0/9.0, Strength=0.22, Dampening=0.82
  276.  
  277. [Running Abstractive Planning...]
  278. Generated 32 abstract experiences (dreams).
  279.  
  280. [Training Models...]
  281. [dBI DEBUG as66-821a4dcad9c2] Total history: 10, Valid BIs: 10
  282. [dBI DEBUG] Last 5 BIs: ['0.1065', '0.1090', '0.0846', '0.1647', '0.1388']
  283. [dBI DEBUG] dbi_raw = 0.138787 - 0.164687 = -0.025900
  284.  
  285. [Training from Abstract Experiences...]
  286. [RETRY] No new levels beaten. Best remains: 2/9
  287. [RETRY] Attempt 2/15. Retrying...
  288.  
  289. ================================================================================
  290. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 3/15
  291. ================================================================================
  292. [INITIAL SCORE] API started at score=2, initializing levels_completed=2
  293. [Ep0] EXPLORATION: No memory replay
  294. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  295. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  296. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  297. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  298. [Ep0] NOT STORED: No level progress (reward=2.00)
  299. [CRYSTALLIZATION] Best: 2.0/9.0, Strength=0.22, Dampening=0.82
  300.  
  301. [Running Abstractive Planning...]
  302. Generated 32 abstract experiences (dreams).
  303.  
  304. [Training Models...]
  305. [dBI DEBUG as66-821a4dcad9c2] Total history: 11, Valid BIs: 11
  306. [dBI DEBUG] Last 5 BIs: ['0.1090', '0.0846', '0.1647', '0.1388', '0.0920']
  307. [dBI DEBUG] dbi_raw = 0.091963 - 0.138787 = -0.046824
  308.  
  309. [Training from Abstract Experiences...]
  310. [RETRY] No new levels beaten. Best remains: 2/9
  311. [RETRY] Attempt 3/15. Retrying...
  312.  
  313. ================================================================================
  314. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 4/15
  315. ================================================================================
  316. [INITIAL SCORE] API started at score=2, initializing levels_completed=2
  317. [Ep0] EXPLORATION: No memory replay
  318. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  319. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  320. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  321. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  322. [Ep0] NOT STORED: No level progress (reward=2.00)
  323. [CRYSTALLIZATION] Best: 2.0/9.0, Strength=0.22, Dampening=0.82
  324.  
  325. [Running Abstractive Planning...]
  326. Generated 32 abstract experiences (dreams).
  327.  
  328. [Training Models...]
  329. [dBI DEBUG as66-821a4dcad9c2] Total history: 12, Valid BIs: 12
  330. [dBI DEBUG] Last 5 BIs: ['0.0846', '0.1647', '0.1388', '0.0920', '0.0769']
  331. [dBI DEBUG] dbi_raw = 0.076917 - 0.091963 = -0.015046
  332.  
  333. [Training from Abstract Experiences...]
  334. [RETRY] No new levels beaten. Best remains: 2/9
  335. [RETRY] Attempt 4/15. Retrying...
  336.  
  337. ================================================================================
  338. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 5/15
  339. ================================================================================
  340. [INITIAL SCORE] API started at score=2, initializing levels_completed=2
  341. [Ep0] EXPLORATION: No memory replay
  342. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  343. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  344. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  345. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  346. [Ep0] NOT STORED: No level progress (reward=2.00)
  347. [CRYSTALLIZATION] Best: 2.0/9.0, Strength=0.22, Dampening=0.82
  348.  
  349. [Running Abstractive Planning...]
  350. Generated 32 abstract experiences (dreams).
  351.  
  352. [Training Models...]
  353. [dBI DEBUG as66-821a4dcad9c2] Total history: 13, Valid BIs: 13
  354. [dBI DEBUG] Last 5 BIs: ['0.1647', '0.1388', '0.0920', '0.0769', '0.1244']
  355. [dBI DEBUG] dbi_raw = 0.124397 - 0.076917 = 0.047480
  356.  
  357. [Training from Abstract Experiences...]
  358. [RETRY] No new levels beaten. Best remains: 2/9
  359. [RETRY] Attempt 5/15. Retrying...
  360.  
  361. ================================================================================
  362. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 6/15
  363. ================================================================================
  364. [INITIAL SCORE] API started at score=2, initializing levels_completed=2
  365. [Ep0] EXPLORATION: No memory replay
  366. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  367. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  368. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  369. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  370. [Ep0] NOT STORED: No level progress (reward=2.00)
  371. [CRYSTALLIZATION] Best: 2.0/9.0, Strength=0.22, Dampening=0.82
  372.  
  373. [Running Abstractive Planning...]
  374. Generated 32 abstract experiences (dreams).
  375.  
  376. [Training Models...]
  377. [dBI DEBUG as66-821a4dcad9c2] Total history: 14, Valid BIs: 14
  378. [dBI DEBUG] Last 5 BIs: ['0.1388', '0.0920', '0.0769', '0.1244', '0.1167']
  379. [dBI DEBUG] dbi_raw = 0.116657 - 0.124397 = -0.007740
  380.  
  381. [Training from Abstract Experiences...]
  382. [RETRY] No new levels beaten. Best remains: 2/9
  383. [RETRY] Attempt 6/15. Retrying...
  384.  
  385. ================================================================================
  386. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 7/15
  387. ================================================================================
  388. [INITIAL SCORE] API started at score=2, initializing levels_completed=2
  389. [Ep0] EXPLORATION: No memory replay
  390. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  391. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  392. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  393. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  394. [Ep0] NOT STORED: No level progress (reward=2.00)
  395. [CRYSTALLIZATION] Best: 2.0/9.0, Strength=0.22, Dampening=0.82
  396.  
  397. [Running Abstractive Planning...]
  398. Generated 32 abstract experiences (dreams).
  399.  
  400. [Training Models...]
  401. [dBI DEBUG as66-821a4dcad9c2] Total history: 15, Valid BIs: 15
  402. [dBI DEBUG] Last 5 BIs: ['0.0920', '0.0769', '0.1244', '0.1167', '0.1243']
  403. [dBI DEBUG] dbi_raw = 0.124306 - 0.116657 = 0.007648
  404.  
  405. [Training from Abstract Experiences...]
  406. [RETRY] No new levels beaten. Best remains: 2/9
  407. [RETRY] Attempt 7/15. Retrying...
  408.  
  409. ================================================================================
  410. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 8/15
  411. ================================================================================
  412. [INITIAL SCORE] API started at score=2, initializing levels_completed=2
  413. [Ep0] EXPLORATION: No memory replay
  414. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  415. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  416. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  417. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  418. [API REWARD] score: 2 -> 3, reward: 1.00
  419. [LEVEL 3 COMPLETE @ step 20] Episode reward: 3.0
  420. [CRYSTALLIZATION] Game as66-821a4dcad9c2: NEW BEST 3.0/9.0 (strength: 0.33)
  421. [CRYSTALLIZATION] Best: 3.0/9.0, Strength=0.33, Dampening=0.73
  422. [SAVED] crystallization.pt updated (new best: 3.0)
  423.  
  424. [Running Abstractive Planning...]
  425. Generated 32 abstract experiences (dreams).
  426.  
  427. [Training Models...]
  428. [dBI DEBUG as66-821a4dcad9c2] Total history: 16, Valid BIs: 16
  429. [dBI DEBUG] Last 5 BIs: ['0.0769', '0.1244', '0.1167', '0.1243', '0.1961']
  430. [dBI DEBUG] dbi_raw = 0.196069 - 0.124306 = 0.071763
  431.  
  432. [Training from Abstract Experiences...]
  433. [RETRY] NEW PROGRESS! Reached 1 NEW levels. Best: 2 -> 3/9
  434. [RETRY] Resetting to Attempt 1/15 due to progress. Retrying...
  435.  
  436. ================================================================================
  437. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 1/15
  438. ================================================================================
  439. [INITIAL SCORE] API started at score=3, initializing levels_completed=3
  440. [Ep0] EXPLORATION: No memory replay
  441. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  442. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  443. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  444. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  445. [API REWARD] score: 3 -> 4, reward: 1.00
  446. [LEVEL 4 COMPLETE @ step 26] Episode reward: 4.0
  447. [CRYSTALLIZATION] Game as66-821a4dcad9c2: NEW BEST 4.0/9.0 (strength: 0.44)
  448. [CRYSTALLIZATION] Best: 4.0/9.0, Strength=0.44, Dampening=0.64
  449. [SAVED] crystallization.pt updated (new best: 4.0)
  450.  
  451. [Running Abstractive Planning...]
  452. Generated 32 abstract experiences (dreams).
  453.  
  454. [Training Models...]
  455. [dBI DEBUG as66-821a4dcad9c2] Total history: 17, Valid BIs: 17
  456. [dBI DEBUG] Last 5 BIs: ['0.1244', '0.1167', '0.1243', '0.1961', '0.1432']
  457. [dBI DEBUG] dbi_raw = 0.143177 - 0.196069 = -0.052892
  458.  
  459. [Training from Abstract Experiences...]
  460. [RETRY] NEW PROGRESS! Reached 1 NEW levels. Best: 3 -> 4/9
  461. [RETRY] Resetting to Attempt 1/15 due to progress. Retrying...
  462.  
  463. ================================================================================
  464. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 1/15
  465. ================================================================================
  466. [INITIAL SCORE] API started at score=4, initializing levels_completed=4
  467. [Ep0] EXPLORATION: No memory replay
  468. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  469. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  470. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  471. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  472. [Ep0] NOT STORED: No level progress (reward=4.00)
  473. [CRYSTALLIZATION] Best: 4.0/9.0, Strength=0.44, Dampening=0.64
  474.  
  475. [Running Abstractive Planning...]
  476. Generated 32 abstract experiences (dreams).
  477.  
  478. [Training Models...]
  479. [dBI DEBUG as66-821a4dcad9c2] Total history: 18, Valid BIs: 18
  480. [dBI DEBUG] Last 5 BIs: ['0.1167', '0.1243', '0.1961', '0.1432', '0.1530']
  481. [dBI DEBUG] dbi_raw = 0.153047 - 0.143177 = 0.009871
  482.  
  483. [Training from Abstract Experiences...]
  484. [RETRY] No new levels beaten. Best remains: 4/9
  485. [RETRY] Attempt 1/15. Retrying...
  486.  
  487. ================================================================================
  488. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 2/15
  489. ================================================================================
  490. [INITIAL SCORE] API started at score=4, initializing levels_completed=4
  491. [Ep0] EXPLORATION: No memory replay
  492. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  493. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  494. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  495. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  496. [Ep0] NOT STORED: No level progress (reward=4.00)
  497. [CRYSTALLIZATION] Best: 4.0/9.0, Strength=0.44, Dampening=0.64
  498.  
  499. [Running Abstractive Planning...]
  500. Generated 32 abstract experiences (dreams).
  501.  
  502. [Training Models...]
  503. [dBI DEBUG as66-821a4dcad9c2] Total history: 19, Valid BIs: 19
  504. [dBI DEBUG] Last 5 BIs: ['0.1243', '0.1961', '0.1432', '0.1530', '0.1501']
  505. [dBI DEBUG] dbi_raw = 0.150057 - 0.153047 = -0.002991
  506.  
  507. [Training from Abstract Experiences...]
  508. [RETRY] No new levels beaten. Best remains: 4/9
  509. [RETRY] Attempt 2/15. Retrying...
  510.  
  511. ================================================================================
  512. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 3/15
  513. ================================================================================
  514. [INITIAL SCORE] API started at score=4, initializing levels_completed=4
  515. [Ep0] EXPLORATION: No memory replay
  516. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  517. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  518. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  519. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  520. [Ep0] NOT STORED: No level progress (reward=4.00)
  521. [CRYSTALLIZATION] Best: 4.0/9.0, Strength=0.44, Dampening=0.64
  522.  
  523. [Running Abstractive Planning...]
  524. Generated 32 abstract experiences (dreams).
  525.  
  526. [Training Models...]
  527. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  528. [dBI DEBUG] Last 5 BIs: ['0.1961', '0.1432', '0.1530', '0.1501', '0.1669']
  529. [dBI DEBUG] dbi_raw = 0.166927 - 0.150057 = 0.016870
  530.  
  531. [Training from Abstract Experiences...]
  532. [RETRY] No new levels beaten. Best remains: 4/9
  533. [RETRY] Attempt 3/15. Retrying...
  534.  
  535. ================================================================================
  536. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 4/15
  537. ================================================================================
  538. [INITIAL SCORE] API started at score=4, initializing levels_completed=4
  539. [Ep0] EXPLORATION: No memory replay
  540. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  541. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  542. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  543. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  544. [Ep0] NOT STORED: No level progress (reward=4.00)
  545. [CRYSTALLIZATION] Best: 4.0/9.0, Strength=0.44, Dampening=0.64
  546.  
  547. [Running Abstractive Planning...]
  548. Generated 32 abstract experiences (dreams).
  549.  
  550. [Training Models...]
  551. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  552. [dBI DEBUG] Last 5 BIs: ['0.1432', '0.1530', '0.1501', '0.1669', '0.1649']
  553. [dBI DEBUG] dbi_raw = 0.164945 - 0.166927 = -0.001982
  554.  
  555. [Training from Abstract Experiences...]
  556. [RETRY] No new levels beaten. Best remains: 4/9
  557. [RETRY] Attempt 4/15. Retrying...
  558.  
  559. ================================================================================
  560. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 5/15
  561. ================================================================================
  562. [INITIAL SCORE] API started at score=4, initializing levels_completed=4
  563. [Ep0] EXPLORATION: No memory replay
  564. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  565. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  566. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  567. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  568. [Ep0] NOT STORED: No level progress (reward=4.00)
  569. [CRYSTALLIZATION] Best: 4.0/9.0, Strength=0.44, Dampening=0.64
  570.  
  571. [Running Abstractive Planning...]
  572. Generated 32 abstract experiences (dreams).
  573.  
  574. [Training Models...]
  575. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  576. [dBI DEBUG] Last 5 BIs: ['0.1530', '0.1501', '0.1669', '0.1649', '0.1431']
  577. [dBI DEBUG] dbi_raw = 0.143101 - 0.164945 = -0.021845
  578.  
  579. [Training from Abstract Experiences...]
  580. [RETRY] No new levels beaten. Best remains: 4/9
  581. [RETRY] Attempt 5/15. Retrying...
  582.  
  583. ================================================================================
  584. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 6/15
  585. ================================================================================
  586. [INITIAL SCORE] API started at score=4, initializing levels_completed=4
  587. [Ep0] EXPLORATION: No memory replay
  588. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  589. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  590. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  591. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  592. [Ep0] NOT STORED: No level progress (reward=4.00)
  593. [CRYSTALLIZATION] Best: 4.0/9.0, Strength=0.44, Dampening=0.64
  594.  
  595. [Running Abstractive Planning...]
  596. Generated 32 abstract experiences (dreams).
  597.  
  598. [Training Models...]
  599. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  600. [dBI DEBUG] Last 5 BIs: ['0.1501', '0.1669', '0.1649', '0.1431', '0.2021']
  601. [dBI DEBUG] dbi_raw = 0.202140 - 0.143101 = 0.059039
  602.  
  603. [Training from Abstract Experiences...]
  604. [RETRY] No new levels beaten. Best remains: 4/9
  605. [RETRY] Attempt 6/15. Retrying...
  606.  
  607. ================================================================================
  608. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 7/15
  609. ================================================================================
  610. [INITIAL SCORE] API started at score=4, initializing levels_completed=4
  611. [Ep0] EXPLORATION: No memory replay
  612. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  613. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  614. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  615. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  616. [Ep0] NOT STORED: No level progress (reward=4.00)
  617. [CRYSTALLIZATION] Best: 4.0/9.0, Strength=0.44, Dampening=0.64
  618.  
  619. [Running Abstractive Planning...]
  620. Generated 32 abstract experiences (dreams).
  621.  
  622. [Training Models...]
  623. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  624. [dBI DEBUG] Last 5 BIs: ['0.1669', '0.1649', '0.1431', '0.2021', '0.1598']
  625. [dBI DEBUG] dbi_raw = 0.159752 - 0.202140 = -0.042388
  626.  
  627. [Training from Abstract Experiences...]
  628. [RETRY] No new levels beaten. Best remains: 4/9
  629. [RETRY] Attempt 7/15. Retrying...
  630.  
  631. ================================================================================
  632. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 8/15
  633. ================================================================================
  634. [INITIAL SCORE] API started at score=4, initializing levels_completed=4
  635. [Ep0] EXPLORATION: No memory replay
  636. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  637. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  638. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  639. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  640. [Ep0] NOT STORED: No level progress (reward=4.00)
  641. [CRYSTALLIZATION] Best: 4.0/9.0, Strength=0.44, Dampening=0.64
  642.  
  643. [Running Abstractive Planning...]
  644. Generated 32 abstract experiences (dreams).
  645.  
  646. [Training Models...]
  647. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  648. [dBI DEBUG] Last 5 BIs: ['0.1649', '0.1431', '0.2021', '0.1598', '0.1673']
  649. [dBI DEBUG] dbi_raw = 0.167327 - 0.159752 = 0.007575
  650.  
  651. [Training from Abstract Experiences...]
  652. [RETRY] No new levels beaten. Best remains: 4/9
  653. [RETRY] Attempt 8/15. Retrying...
  654.  
  655. ================================================================================
  656. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 9/15
  657. ================================================================================
  658. [INITIAL SCORE] API started at score=4, initializing levels_completed=4
  659. [Ep0] EXPLORATION: No memory replay
  660. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  661. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  662. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  663. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  664. [Ep0] NOT STORED: No level progress (reward=4.00)
  665. [CRYSTALLIZATION] Best: 4.0/9.0, Strength=0.44, Dampening=0.64
  666.  
  667. [Running Abstractive Planning...]
  668. Generated 32 abstract experiences (dreams).
  669.  
  670. [Training Models...]
  671. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  672. [dBI DEBUG] Last 5 BIs: ['0.1431', '0.2021', '0.1598', '0.1673', '0.1786']
  673. [dBI DEBUG] dbi_raw = 0.178576 - 0.167327 = 0.011249
  674.  
  675. [Training from Abstract Experiences...]
  676. [RETRY] No new levels beaten. Best remains: 4/9
  677. [RETRY] Attempt 9/15. Retrying...
  678.  
  679. ================================================================================
  680. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 10/15
  681. ================================================================================
  682. [INITIAL SCORE] API started at score=4, initializing levels_completed=4
  683. [Ep0] EXPLORATION: No memory replay
  684. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  685. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  686. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  687. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  688. [Ep0] NOT STORED: No level progress (reward=4.00)
  689. [CRYSTALLIZATION] Best: 4.0/9.0, Strength=0.44, Dampening=0.64
  690.  
  691. [Running Abstractive Planning...]
  692. Generated 32 abstract experiences (dreams).
  693.  
  694. [Training Models...]
  695. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  696. [dBI DEBUG] Last 5 BIs: ['0.2021', '0.1598', '0.1673', '0.1786', '0.1760']
  697. [dBI DEBUG] dbi_raw = 0.175956 - 0.178576 = -0.002620
  698.  
  699. [Training from Abstract Experiences...]
  700. [RETRY] No new levels beaten. Best remains: 4/9
  701. [RETRY] Attempt 10/15. Retrying...
  702.  
  703. ================================================================================
  704. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 11/15
  705. ================================================================================
  706. [INITIAL SCORE] API started at score=4, initializing levels_completed=4
  707. [Ep0] EXPLORATION: No memory replay
  708. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  709. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  710. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  711. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  712. [API REWARD] score: 4 -> 5, reward: 1.00
  713. [LEVEL 5 COMPLETE @ step 69] Episode reward: 5.0
  714. [CRYSTALLIZATION] Game as66-821a4dcad9c2: NEW BEST 5.0/9.0 (strength: 0.56)
  715. [CRYSTALLIZATION] Best: 5.0/9.0, Strength=0.56, Dampening=0.56
  716. [SAVED] crystallization.pt updated (new best: 5.0)
  717.  
  718. [Running Abstractive Planning...]
  719. Generated 32 abstract experiences (dreams).
  720.  
  721. [Training Models...]
  722. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  723. [dBI DEBUG] Last 5 BIs: ['0.1598', '0.1673', '0.1786', '0.1760', '0.2006']
  724. [dBI DEBUG] dbi_raw = 0.200621 - 0.175956 = 0.024664
  725.  
  726. [Training from Abstract Experiences...]
  727. [RETRY] NEW PROGRESS! Reached 1 NEW levels. Best: 4 -> 5/9
  728. [RETRY] Resetting to Attempt 1/15 due to progress. Retrying...
  729.  
  730. ================================================================================
  731. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 1/15
  732. ================================================================================
  733. [INITIAL SCORE] API started at score=5, initializing levels_completed=5
  734. [Ep0] EXPLORATION: No memory replay
  735. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  736. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  737. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  738. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  739. [Ep0] NOT STORED: No level progress (reward=5.00)
  740. [CRYSTALLIZATION] Best: 5.0/9.0, Strength=0.56, Dampening=0.56
  741.  
  742. [Running Abstractive Planning...]
  743. Generated 32 abstract experiences (dreams).
  744.  
  745. [Training Models...]
  746. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  747. [dBI DEBUG] Last 5 BIs: ['0.1673', '0.1786', '0.1760', '0.2006', '0.1348']
  748. [dBI DEBUG] dbi_raw = 0.134790 - 0.200621 = -0.065831
  749.  
  750. [Training from Abstract Experiences...]
  751. [RETRY] No new levels beaten. Best remains: 5/9
  752. [RETRY] Attempt 1/15. Retrying...
  753.  
  754. ================================================================================
  755. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 2/15
  756. ================================================================================
  757. [INITIAL SCORE] API started at score=5, initializing levels_completed=5
  758. [Ep0] EXPLORATION: No memory replay
  759. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  760. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  761. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  762. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  763. [Ep0] NOT STORED: No level progress (reward=5.00)
  764. [CRYSTALLIZATION] Best: 5.0/9.0, Strength=0.56, Dampening=0.56
  765.  
  766. [Running Abstractive Planning...]
  767. Generated 32 abstract experiences (dreams).
  768.  
  769. [Training Models...]
  770. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  771. [dBI DEBUG] Last 5 BIs: ['0.1786', '0.1760', '0.2006', '0.1348', '0.2048']
  772. [dBI DEBUG] dbi_raw = 0.204762 - 0.134790 = 0.069971
  773.  
  774. [Training from Abstract Experiences...]
  775. [RETRY] No new levels beaten. Best remains: 5/9
  776. [RETRY] Attempt 2/15. Retrying...
  777.  
  778. ================================================================================
  779. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 3/15
  780. ================================================================================
  781. [INITIAL SCORE] API started at score=5, initializing levels_completed=5
  782. [Ep0] EXPLORATION: No memory replay
  783. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  784. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  785. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  786. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  787. [Ep0] NOT STORED: No level progress (reward=5.00)
  788. [CRYSTALLIZATION] Best: 5.0/9.0, Strength=0.56, Dampening=0.56
  789.  
  790. [Running Abstractive Planning...]
  791. Generated 32 abstract experiences (dreams).
  792.  
  793. [Training Models...]
  794. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  795. [dBI DEBUG] Last 5 BIs: ['0.1760', '0.2006', '0.1348', '0.2048', '0.1584']
  796. [dBI DEBUG] dbi_raw = 0.158408 - 0.204762 = -0.046354
  797.  
  798. [Training from Abstract Experiences...]
  799. [RETRY] No new levels beaten. Best remains: 5/9
  800. [RETRY] Attempt 3/15. Retrying...
  801.  
  802. ================================================================================
  803. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 4/15
  804. ================================================================================
  805. [INITIAL SCORE] API started at score=5, initializing levels_completed=5
  806. [Ep0] EXPLORATION: No memory replay
  807. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  808. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  809. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  810. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  811. [Ep0] NOT STORED: No level progress (reward=5.00)
  812. [CRYSTALLIZATION] Best: 5.0/9.0, Strength=0.56, Dampening=0.56
  813.  
  814. [Running Abstractive Planning...]
  815. Generated 32 abstract experiences (dreams).
  816.  
  817. [Training Models...]
  818. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  819. [dBI DEBUG] Last 5 BIs: ['0.2006', '0.1348', '0.2048', '0.1584', '0.1888']
  820. [dBI DEBUG] dbi_raw = 0.188766 - 0.158408 = 0.030358
  821.  
  822. [Training from Abstract Experiences...]
  823. [RETRY] No new levels beaten. Best remains: 5/9
  824. [RETRY] Attempt 4/15. Retrying...
  825.  
  826. ================================================================================
  827. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 5/15
  828. ================================================================================
  829. [INITIAL SCORE] API started at score=5, initializing levels_completed=5
  830. [Ep0] EXPLORATION: No memory replay
  831. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  832. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  833. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  834. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  835. [Ep0] NOT STORED: No level progress (reward=5.00)
  836. [CRYSTALLIZATION] Best: 5.0/9.0, Strength=0.56, Dampening=0.56
  837.  
  838. [Running Abstractive Planning...]
  839. Generated 32 abstract experiences (dreams).
  840.  
  841. [Training Models...]
  842. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  843. [dBI DEBUG] Last 5 BIs: ['0.1348', '0.2048', '0.1584', '0.1888', '0.2095']
  844. [dBI DEBUG] dbi_raw = 0.209511 - 0.188766 = 0.020745
  845.  
  846. [Training from Abstract Experiences...]
  847. Abstract Value Loss: 1427.1710
  848. Abstract Policy Loss: 1.8204
  849. [RETRY] No new levels beaten. Best remains: 5/9
  850. [RETRY] Attempt 5/15. Retrying...
  851.  
  852. ================================================================================
  853. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 6/15
  854. ================================================================================
  855. [INITIAL SCORE] API started at score=5, initializing levels_completed=5
  856. [Ep0] EXPLORATION: No memory replay
  857. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  858. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  859. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  860. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  861. [API REWARD] score: 5 -> 6, reward: 1.00
  862. [LEVEL 6 COMPLETE @ step 15] Episode reward: 6.0
  863. [CRYSTALLIZATION] Game as66-821a4dcad9c2: NEW BEST 6.0/9.0 (strength: 0.67)
  864. [CRYSTALLIZATION] Best: 6.0/9.0, Strength=0.67, Dampening=0.47
  865. [SAVED] crystallization.pt updated (new best: 6.0)
  866.  
  867. [Running Abstractive Planning...]
  868. Generated 32 abstract experiences (dreams).
  869.  
  870. [Training Models...]
  871. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  872. [dBI DEBUG] Last 5 BIs: ['0.2048', '0.1584', '0.1888', '0.2095', '0.1642']
  873. [dBI DEBUG] dbi_raw = 0.164205 - 0.209511 = -0.045306
  874.  
  875. [Training from Abstract Experiences...]
  876. [RETRY] NEW PROGRESS! Reached 1 NEW levels. Best: 5 -> 6/9
  877. [RETRY] Resetting to Attempt 1/15 due to progress. Retrying...
  878.  
  879. ================================================================================
  880. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 1/15
  881. ================================================================================
  882. [INITIAL SCORE] API started at score=6, initializing levels_completed=6
  883. [Ep0] EXPLORATION: No memory replay
  884. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  885. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  886. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  887. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  888. [Ep0] NOT STORED: No level progress (reward=6.00)
  889. [CRYSTALLIZATION] Best: 6.0/9.0, Strength=0.67, Dampening=0.47
  890.  
  891. [Running Abstractive Planning...]
  892. Generated 32 abstract experiences (dreams).
  893.  
  894. [Training Models...]
  895. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  896. [dBI DEBUG] Last 5 BIs: ['0.1584', '0.1888', '0.2095', '0.1642', '0.2483']
  897. [dBI DEBUG] dbi_raw = 0.248301 - 0.164205 = 0.084096
  898.  
  899. [Training from Abstract Experiences...]
  900. [RETRY] No new levels beaten. Best remains: 6/9
  901. [RETRY] Attempt 1/15. Retrying...
  902.  
  903. ================================================================================
  904. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 2/15
  905. ================================================================================
  906. [INITIAL SCORE] API started at score=6, initializing levels_completed=6
  907. [Ep0] EXPLORATION: No memory replay
  908. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  909. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  910. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  911. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  912. [API REWARD] score: 6 -> 7, reward: 1.00
  913. [LEVEL 7 COMPLETE @ step 78] Episode reward: 7.0
  914. [CRYSTALLIZATION] Game as66-821a4dcad9c2: NEW BEST 7.0/9.0 (strength: 0.78)
  915. [CRYSTALLIZATION] Best: 7.0/9.0, Strength=0.78, Dampening=0.38
  916. [SAVED] crystallization.pt updated (new best: 7.0)
  917.  
  918. [Running Abstractive Planning...]
  919. Generated 32 abstract experiences (dreams).
  920.  
  921. [Training Models...]
  922. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  923. [dBI DEBUG] Last 5 BIs: ['0.1888', '0.2095', '0.1642', '0.2483', '0.1501']
  924. [dBI DEBUG] dbi_raw = 0.150085 - 0.248301 = -0.098216
  925.  
  926. [Training from Abstract Experiences...]
  927. [RETRY] NEW PROGRESS! Reached 1 NEW levels. Best: 6 -> 7/9
  928. [RETRY] Resetting to Attempt 1/15 due to progress. Retrying...
  929.  
  930. ================================================================================
  931. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 1/15
  932. ================================================================================
  933. [INITIAL SCORE] API started at score=7, initializing levels_completed=7
  934. [Ep0] EXPLORATION: No memory replay
  935. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  936. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  937. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  938. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  939. [Ep0] NOT STORED: No level progress (reward=7.00)
  940. [CRYSTALLIZATION] Best: 7.0/9.0, Strength=0.78, Dampening=0.38
  941.  
  942. [Running Abstractive Planning...]
  943. Generated 32 abstract experiences (dreams).
  944.  
  945. [Training Models...]
  946. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  947. [dBI DEBUG] Last 5 BIs: ['0.2095', '0.1642', '0.2483', '0.1501', '0.2287']
  948. [dBI DEBUG] dbi_raw = 0.228709 - 0.150085 = 0.078624
  949.  
  950. [Training from Abstract Experiences...]
  951. Abstract Value Loss: 269.2734
  952. Abstract Policy Loss: 1.8474
  953. [RETRY] No new levels beaten. Best remains: 7/9
  954. [RETRY] Attempt 1/15. Retrying...
  955.  
  956. ================================================================================
  957. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 2/15
  958. ================================================================================
  959. [INITIAL SCORE] API started at score=7, initializing levels_completed=7
  960. [Ep0] EXPLORATION: No memory replay
  961. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  962. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  963. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  964. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  965. [Ep0] NOT STORED: No level progress (reward=7.00)
  966. [CRYSTALLIZATION] Best: 7.0/9.0, Strength=0.78, Dampening=0.38
  967.  
  968. [Running Abstractive Planning...]
  969. Generated 32 abstract experiences (dreams).
  970.  
  971. [Training Models...]
  972. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  973. [dBI DEBUG] Last 5 BIs: ['0.1642', '0.2483', '0.1501', '0.2287', '0.1417']
  974. [dBI DEBUG] dbi_raw = 0.141687 - 0.228709 = -0.087023
  975.  
  976. [Training from Abstract Experiences...]
  977. [RETRY] No new levels beaten. Best remains: 7/9
  978. [RETRY] Attempt 2/15. Retrying...
  979.  
  980. ================================================================================
  981. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 3/15
  982. ================================================================================
  983. [INITIAL SCORE] API started at score=7, initializing levels_completed=7
  984. [Ep0] EXPLORATION: No memory replay
  985. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  986. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  987. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  988. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  989. [Ep0] NOT STORED: No level progress (reward=7.00)
  990. [CRYSTALLIZATION] Best: 7.0/9.0, Strength=0.78, Dampening=0.38
  991.  
  992. [Running Abstractive Planning...]
  993. Generated 32 abstract experiences (dreams).
  994.  
  995. [Training Models...]
  996. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  997. [dBI DEBUG] Last 5 BIs: ['0.2483', '0.1501', '0.2287', '0.1417', '0.1514']
  998. [dBI DEBUG] dbi_raw = 0.151352 - 0.141687 = 0.009665
  999.  
  1000. [Training from Abstract Experiences...]
  1001. [RETRY] No new levels beaten. Best remains: 7/9
  1002. [RETRY] Attempt 3/15. Retrying...
  1003.  
  1004. ================================================================================
  1005. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 4/15
  1006. ================================================================================
  1007. [INITIAL SCORE] API started at score=7, initializing levels_completed=7
  1008. [Ep0] EXPLORATION: No memory replay
  1009. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  1010. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1011. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  1012. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1013. [Ep0] NOT STORED: No level progress (reward=7.00)
  1014. [CRYSTALLIZATION] Best: 7.0/9.0, Strength=0.78, Dampening=0.38
  1015.  
  1016. [Running Abstractive Planning...]
  1017. Generated 32 abstract experiences (dreams).
  1018.  
  1019. [Training Models...]
  1020. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  1021. [dBI DEBUG] Last 5 BIs: ['0.1501', '0.2287', '0.1417', '0.1514', '0.1695']
  1022. [dBI DEBUG] dbi_raw = 0.169455 - 0.151352 = 0.018103
  1023.  
  1024. [Training from Abstract Experiences...]
  1025. [RETRY] No new levels beaten. Best remains: 7/9
  1026. [RETRY] Attempt 4/15. Retrying...
  1027.  
  1028. ================================================================================
  1029. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 5/15
  1030. ================================================================================
  1031. [INITIAL SCORE] API started at score=7, initializing levels_completed=7
  1032. [Ep0] EXPLORATION: No memory replay
  1033. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  1034. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1035. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  1036. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1037. [Ep0] NOT STORED: No level progress (reward=7.00)
  1038. [CRYSTALLIZATION] Best: 7.0/9.0, Strength=0.78, Dampening=0.38
  1039.  
  1040. [Running Abstractive Planning...]
  1041. Generated 32 abstract experiences (dreams).
  1042.  
  1043. [Training Models...]
  1044. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  1045. [dBI DEBUG] Last 5 BIs: ['0.2287', '0.1417', '0.1514', '0.1695', '0.1718']
  1046. [dBI DEBUG] dbi_raw = 0.171847 - 0.169455 = 0.002392
  1047.  
  1048. [Training from Abstract Experiences...]
  1049. [RETRY] No new levels beaten. Best remains: 7/9
  1050. [RETRY] Attempt 5/15. Retrying...
  1051.  
  1052. ================================================================================
  1053. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 6/15
  1054. ================================================================================
  1055. [INITIAL SCORE] API started at score=7, initializing levels_completed=7
  1056. [Ep0] EXPLORATION: No memory replay
  1057. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  1058. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1059. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  1060. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1061. [Ep0] NOT STORED: No level progress (reward=7.00)
  1062. [CRYSTALLIZATION] Best: 7.0/9.0, Strength=0.78, Dampening=0.38
  1063.  
  1064. [Running Abstractive Planning...]
  1065. Generated 32 abstract experiences (dreams).
  1066.  
  1067. [Training Models...]
  1068. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  1069. [dBI DEBUG] Last 5 BIs: ['0.1417', '0.1514', '0.1695', '0.1718', '0.1584']
  1070. [dBI DEBUG] dbi_raw = 0.158435 - 0.171847 = -0.013411
  1071.  
  1072. [Training from Abstract Experiences...]
  1073. [RETRY] No new levels beaten. Best remains: 7/9
  1074. [RETRY] Attempt 6/15. Retrying...
  1075.  
  1076. ================================================================================
  1077. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 7/15
  1078. ================================================================================
  1079. [INITIAL SCORE] API started at score=7, initializing levels_completed=7
  1080. [Ep0] EXPLORATION: No memory replay
  1081. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  1082. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1083. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  1084. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1085. [Ep0] NOT STORED: No level progress (reward=7.00)
  1086. [CRYSTALLIZATION] Best: 7.0/9.0, Strength=0.78, Dampening=0.38
  1087.  
  1088. [Running Abstractive Planning...]
  1089. Generated 32 abstract experiences (dreams).
  1090.  
  1091. [Training Models...]
  1092. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  1093. [dBI DEBUG] Last 5 BIs: ['0.1514', '0.1695', '0.1718', '0.1584', '0.1595']
  1094. [dBI DEBUG] dbi_raw = 0.159509 - 0.158435 = 0.001073
  1095.  
  1096. [Training from Abstract Experiences...]
  1097. [RETRY] No new levels beaten. Best remains: 7/9
  1098. [RETRY] Attempt 7/15. Retrying...
  1099.  
  1100. ================================================================================
  1101. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 8/15
  1102. ================================================================================
  1103. [INITIAL SCORE] API started at score=7, initializing levels_completed=7
  1104. [Ep0] EXPLORATION: No memory replay
  1105. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  1106. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1107. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  1108. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1109. [Ep0] NOT STORED: No level progress (reward=7.00)
  1110. [CRYSTALLIZATION] Best: 7.0/9.0, Strength=0.78, Dampening=0.38
  1111.  
  1112. [Running Abstractive Planning...]
  1113. Generated 32 abstract experiences (dreams).
  1114.  
  1115. [Training Models...]
  1116. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  1117. [dBI DEBUG] Last 5 BIs: ['0.1695', '0.1718', '0.1584', '0.1595', '0.2448']
  1118. [dBI DEBUG] dbi_raw = 0.244764 - 0.159509 = 0.085255
  1119.  
  1120. [Training from Abstract Experiences...]
  1121. Abstract Value Loss: 503.8107
  1122. Abstract Policy Loss: 1.8477
  1123. [RETRY] No new levels beaten. Best remains: 7/9
  1124. [RETRY] Attempt 8/15. Retrying...
  1125.  
  1126. ================================================================================
  1127. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 9/15
  1128. ================================================================================
  1129. [INITIAL SCORE] API started at score=7, initializing levels_completed=7
  1130. [Ep0] EXPLORATION: No memory replay
  1131. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  1132. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1133. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  1134. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1135. [Ep0] NOT STORED: No level progress (reward=7.00)
  1136. [CRYSTALLIZATION] Best: 7.0/9.0, Strength=0.78, Dampening=0.38
  1137.  
  1138. [Running Abstractive Planning...]
  1139. Generated 32 abstract experiences (dreams).
  1140.  
  1141. [Training Models...]
  1142. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  1143. [dBI DEBUG] Last 5 BIs: ['0.1718', '0.1584', '0.1595', '0.2448', '0.1383']
  1144. [dBI DEBUG] dbi_raw = 0.138342 - 0.244764 = -0.106421
  1145.  
  1146. [Training from Abstract Experiences...]
  1147. [RETRY] No new levels beaten. Best remains: 7/9
  1148. [RETRY] Attempt 9/15. Retrying...
  1149.  
  1150. ================================================================================
  1151. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 10/15
  1152. ================================================================================
  1153. [INITIAL SCORE] API started at score=7, initializing levels_completed=7
  1154. [Ep0] EXPLORATION: No memory replay
  1155. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  1156. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1157. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  1158. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1159. [Ep0] NOT STORED: No level progress (reward=7.00)
  1160. [CRYSTALLIZATION] Best: 7.0/9.0, Strength=0.78, Dampening=0.38
  1161.  
  1162. [Running Abstractive Planning...]
  1163. Generated 32 abstract experiences (dreams).
  1164.  
  1165. [Training Models...]
  1166. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  1167. [dBI DEBUG] Last 5 BIs: ['0.1584', '0.1595', '0.2448', '0.1383', '0.1833']
  1168. [dBI DEBUG] dbi_raw = 0.183313 - 0.138342 = 0.044971
  1169.  
  1170. [Training from Abstract Experiences...]
  1171. [RETRY] No new levels beaten. Best remains: 7/9
  1172. [RETRY] Attempt 10/15. Retrying...
  1173.  
  1174. ================================================================================
  1175. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 11/15
  1176. ================================================================================
  1177. [INITIAL SCORE] API started at score=7, initializing levels_completed=7
  1178. [Ep0] EXPLORATION: No memory replay
  1179. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  1180. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1181. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  1182. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1183. [Ep0] NOT STORED: No level progress (reward=7.00)
  1184. [CRYSTALLIZATION] Best: 7.0/9.0, Strength=0.78, Dampening=0.38
  1185.  
  1186. [Running Abstractive Planning...]
  1187. Generated 32 abstract experiences (dreams).
  1188.  
  1189. [Training Models...]
  1190. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  1191. [dBI DEBUG] Last 5 BIs: ['0.1595', '0.2448', '0.1383', '0.1833', '0.2034']
  1192. [dBI DEBUG] dbi_raw = 0.203363 - 0.183313 = 0.020050
  1193.  
  1194. [Training from Abstract Experiences...]
  1195. Abstract Value Loss: 302.0228
  1196. Abstract Policy Loss: 1.8496
  1197. [RETRY] No new levels beaten. Best remains: 7/9
  1198. [RETRY] Attempt 11/15. Retrying...
  1199.  
  1200. ================================================================================
  1201. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 12/15
  1202. ================================================================================
  1203. [INITIAL SCORE] API started at score=7, initializing levels_completed=7
  1204. [Ep0] EXPLORATION: No memory replay
  1205. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  1206. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1207. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  1208. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1209. [Ep0] NOT STORED: No level progress (reward=7.00)
  1210. [CRYSTALLIZATION] Best: 7.0/9.0, Strength=0.78, Dampening=0.38
  1211.  
  1212. [Running Abstractive Planning...]
  1213. Generated 32 abstract experiences (dreams).
  1214.  
  1215. [Training Models...]
  1216. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  1217. [dBI DEBUG] Last 5 BIs: ['0.2448', '0.1383', '0.1833', '0.2034', '0.1798']
  1218. [dBI DEBUG] dbi_raw = 0.179812 - 0.203363 = -0.023550
  1219.  
  1220. [Training from Abstract Experiences...]
  1221. [RETRY] No new levels beaten. Best remains: 7/9
  1222. [RETRY] Attempt 12/15. Retrying...
  1223.  
  1224. ================================================================================
  1225. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 13/15
  1226. ================================================================================
  1227. [INITIAL SCORE] API started at score=7, initializing levels_completed=7
  1228. [Ep0] EXPLORATION: No memory replay
  1229. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  1230. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1231. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  1232. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1233. [Ep0] NOT STORED: No level progress (reward=7.00)
  1234. [CRYSTALLIZATION] Best: 7.0/9.0, Strength=0.78, Dampening=0.38
  1235.  
  1236. [Running Abstractive Planning...]
  1237. Generated 32 abstract experiences (dreams).
  1238.  
  1239. [Training Models...]
  1240. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  1241. [dBI DEBUG] Last 5 BIs: ['0.1383', '0.1833', '0.2034', '0.1798', '0.1436']
  1242. [dBI DEBUG] dbi_raw = 0.143637 - 0.179812 = -0.036175
  1243.  
  1244. [Training from Abstract Experiences...]
  1245. [RETRY] No new levels beaten. Best remains: 7/9
  1246. [RETRY] Attempt 13/15. Retrying...
  1247.  
  1248. ================================================================================
  1249. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 14/15
  1250. ================================================================================
  1251. [INITIAL SCORE] API started at score=7, initializing levels_completed=7
  1252. [Ep0] EXPLORATION: No memory replay
  1253. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  1254. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1255. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  1256. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1257. [Ep0] NOT STORED: No level progress (reward=7.00)
  1258. [CRYSTALLIZATION] Best: 7.0/9.0, Strength=0.78, Dampening=0.38
  1259.  
  1260. [Running Abstractive Planning...]
  1261. Generated 32 abstract experiences (dreams).
  1262.  
  1263. [Training Models...]
  1264. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  1265. [dBI DEBUG] Last 5 BIs: ['0.1833', '0.2034', '0.1798', '0.1436', '0.1915']
  1266. [dBI DEBUG] dbi_raw = 0.191460 - 0.143637 = 0.047822
  1267.  
  1268. [Training from Abstract Experiences...]
  1269. Abstract Value Loss: 619.0808
  1270. Abstract Policy Loss: 1.8507
  1271. [RETRY] No new levels beaten. Best remains: 7/9
  1272. [RETRY] Attempt 14/15. Retrying...
  1273.  
  1274. ================================================================================
  1275. [FinalGift 8.0] Episode 0 | Game: as66-821a4dcad9c2 | Attempt 15/15
  1276. ================================================================================
  1277. [INITIAL SCORE] API started at score=7, initializing levels_completed=7
  1278. [Ep0] EXPLORATION: No memory replay
  1279. [Game: as66-821a4dcad9c2] API available_actions: [1, 2, 3, 4, 6]
  1280. [Game: as66-821a4dcad9c2] Action mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1281. [DEBUG] available_actions from API: [1, 2, 3, 4, 6]
  1282. [DEBUG] action_mask: {0: True, 1: True, 2: True, 3: True, 'click': True, 4: False, 5: False}
  1283. [Ep0] NOT STORED: No level progress (reward=7.00)
  1284. [CRYSTALLIZATION] Best: 7.0/9.0, Strength=0.78, Dampening=0.38
  1285.  
  1286. [Running Abstractive Planning...]
  1287. Generated 32 abstract experiences (dreams).
  1288.  
  1289. [Training Models...]
  1290. [dBI DEBUG as66-821a4dcad9c2] Total history: 20, Valid BIs: 20
  1291. [dBI DEBUG] Last 5 BIs: ['0.2034', '0.1798', '0.1436', '0.1915', '0.1753']
  1292. [dBI DEBUG] dbi_raw = 0.175284 - 0.191460 = -0.016175
  1293.  
  1294. [Training from Abstract Experiences...]
  1295. [RETRY] No new levels beaten. Best remains: 7/9
  1296. [RETRY] Max retries (15) reached without improvement. Moving to next game.
  1297. [SCORECARD CLOSED] View replay at: https://three.arcprize.org/scorecards/ff0e097a-e654-4961-bc1c-997d013fd5a9
  1298.  
  1299. Episode Results:
  1300. Reward: 7.000
  1301. Steps: 6
  1302. Mean Confidence: 0.547
  1303. Mean Uncertainty: 0.629
  1304. Mode Counts: {'explore': 6}
  1305.  
  1306. Model Metrics:
  1307. Prediction Error: 0.0038
  1308. Inverse Loss: 0.9070
  1309. Policy Loss: 0.0000
  1310. Info Gain Loss: 0.0000
  1311.  
  1312. Proto-AGI Module Losses:
  1313. Subgoal Value: 0.0052
  1314. Subgoal Proposer: -0.1222
  1315. Subgoal Achievement: 0.0105
  1316. Rule Encoder: -0.4151
  1317. Rule Confidence: 0.6881
  1318. Empowerment Forward: 0.0131
  1319. Empowerment Inverse: 0.9211
  1320. Counterfactual: 0.0011
  1321. MetaLearner Context: 0.0000
  1322.  
  1323. Per-Step Diagnostics (v8.0):
  1324. Learned Value: 0.5530
  1325. H_op (Operator Entropy): 0.9984
  1326. H_attn (Attention Entropy): 0.6308
  1327. MoE Gate Sparsity: 0.0013
  1328. Empowerment Bonus: 0.7371
  1329. GNN Entropy Reg: 0.0000
  1330.  
  1331. MuZero Boost (Planner→Policy Distillation):
  1332. Value Distillation Loss: 0.0094
  1333. Policy Distillation Loss: 1.7919
  1334.  
  1335. Broad Intelligence:
  1336. A_sub: 0.9986
  1337. A_con: 0.4337
  1338. I_alg: 0.4048 (op_H=0.998, attn_H=0.631)
  1339. BI: 0.1753
  1340. CP: 0.1751 (Choice Potential = BI / 1+C)
  1341. dBI: -0.0005 | dBI-EMA: -0.0005 | dBI-norm: -0.191
  1342. Entropy_penalty: 0.0548 (target=0.10, pushes attn_H)
  1343.  
  1344. Execution Blend Controller:
  1345. Execution: DELIBERATE (w=0.429)
  1346. Delta_intuitive: 0.0597
  1347. Delta_deliberate: -0.0597
  1348. Motion CAD: -0.0085
  1349. Intent Ratio: 100.0% LEARN
  1350.  
  1351. Game Statistics:
  1352. as66-821a4dcad9c2: 51 attempts, 98.0% win, 4.37 avg
  1353. ft09-b7249182: 0 attempts, 0.0% win, 0.00 avg
  1354. lp85-e179de251: 0 attempts, 0.0% win, 0.00 avg
  1355. ls20-e98196ab: 0 attempts, 0.0% win, 0.00 avg
  1356. sp80-445eab21: 0 attempts, 0.0% win, 0.00 avg
  1357. vc33-cbded52d: 0 attempts, 0.0% win, 0.00 avg
  1358.  
  1359. ================================================================================
  1360. Cognitive Mode Analysis (Episode 0):
  1361. ================================================================================
  1362. REDACTED ------
Tags: machine learning
Advertisement
Add Comment
Please, Sign In to add comment
Public Pastes
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!