Alex-GSL commited on
Commit
71a3320
·
verified ·
1 Parent(s): c89c176

Upload human_games/game_20260325_182418_6.json with huggingface_hub

Browse files
human_games/game_20260325_182418_6.json ADDED
@@ -0,0 +1,2751 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "game_number": 6,
3
+ "timestamp": "2026-03-25T18:24:18.955519+00:00",
4
+ "checkpoint": "checkpoints/r42_1500M.pkl",
5
+ "architecture": "simba_aux",
6
+ "model_checkpoint": "r42_1500M.pkl",
7
+ "model_architecture": "simba_aux",
8
+ "result": "model_win",
9
+ "final_scores": [
10
+ 0,
11
+ 8
12
+ ],
13
+ "total_turns": 19,
14
+ "actions": [
15
+ {
16
+ "turn": 0,
17
+ "player": -1,
18
+ "phase": "draw",
19
+ "action": -1,
20
+ "action_desc": "Initial deal",
21
+ "card_drawn": null,
22
+ "hands": [
23
+ [
24
+ 6,
25
+ 18,
26
+ 44,
27
+ 32,
28
+ 29,
29
+ 51,
30
+ 1,
31
+ 4,
32
+ 12,
33
+ 49
34
+ ],
35
+ [
36
+ 34,
37
+ 17,
38
+ 37,
39
+ 35,
40
+ 47,
41
+ 39,
42
+ 26,
43
+ 3,
44
+ 10,
45
+ 38
46
+ ]
47
+ ],
48
+ "hand_sizes": [
49
+ 10,
50
+ 10
51
+ ],
52
+ "discard_pile": [
53
+ 50
54
+ ],
55
+ "discard_size": 1,
56
+ "stock_remaining": 31,
57
+ "deadwood": [
58
+ 67,
59
+ 69
60
+ ],
61
+ "model_logits": null,
62
+ "model_value": null,
63
+ "model_checkpoint": "r42_1500M.pkl",
64
+ "model_architecture": "simba_aux"
65
+ },
66
+ {
67
+ "turn": 1,
68
+ "player": 1,
69
+ "phase": "draw",
70
+ "action": 1,
71
+ "action_desc": "Model drew Q\u2663 from discard",
72
+ "card_drawn": 50,
73
+ "hands": [
74
+ [
75
+ 6,
76
+ 18,
77
+ 44,
78
+ 32,
79
+ 29,
80
+ 51,
81
+ 1,
82
+ 4,
83
+ 12,
84
+ 49
85
+ ],
86
+ [
87
+ 34,
88
+ 17,
89
+ 37,
90
+ 35,
91
+ 47,
92
+ 39,
93
+ 26,
94
+ 3,
95
+ 10,
96
+ 38
97
+ ]
98
+ ],
99
+ "hand_sizes": [
100
+ 10,
101
+ 10
102
+ ],
103
+ "discard_pile": [
104
+ 50
105
+ ],
106
+ "discard_size": 1,
107
+ "stock_remaining": 31,
108
+ "deadwood": [
109
+ 67,
110
+ 69
111
+ ],
112
+ "model_logits": [
113
+ -4.4753313064575195,
114
+ 4.474640846252441,
115
+ -7.466315746307373,
116
+ 4.56476354598999,
117
+ -2.559753179550171,
118
+ -7.512016773223877,
119
+ -1.93171226978302,
120
+ 0.060535743832588196,
121
+ 3.0119740962982178,
122
+ 4.870151996612549,
123
+ -2.2620508670806885,
124
+ -0.29784661531448364,
125
+ 1.3456090688705444,
126
+ 0.19386403262615204,
127
+ -1.45075523853302,
128
+ 13.128887176513672
129
+ ],
130
+ "model_value": 0.31912970542907715,
131
+ "model_checkpoint": "r42_1500M.pkl",
132
+ "model_architecture": "simba_aux"
133
+ },
134
+ {
135
+ "turn": 2,
136
+ "player": 1,
137
+ "phase": "discard",
138
+ "action": 9,
139
+ "action_desc": "Model discarded 4\u2660",
140
+ "card_drawn": null,
141
+ "hands": [
142
+ [
143
+ 6,
144
+ 18,
145
+ 44,
146
+ 32,
147
+ 29,
148
+ 51,
149
+ 1,
150
+ 4,
151
+ 12,
152
+ 49
153
+ ],
154
+ [
155
+ 34,
156
+ 17,
157
+ 37,
158
+ 35,
159
+ 47,
160
+ 39,
161
+ 26,
162
+ 3,
163
+ 10,
164
+ 38,
165
+ 50
166
+ ]
167
+ ],
168
+ "hand_sizes": [
169
+ 10,
170
+ 11
171
+ ],
172
+ "discard_pile": [],
173
+ "discard_size": 0,
174
+ "stock_remaining": 31,
175
+ "deadwood": [
176
+ 67,
177
+ 79
178
+ ],
179
+ "model_logits": [
180
+ 1.1842138767242432,
181
+ -1.1893731355667114,
182
+ -2.9357717037200928,
183
+ 5.730153560638428,
184
+ -9.111689567565918,
185
+ -5.837414264678955,
186
+ 0.392707884311676,
187
+ 4.241512775421143,
188
+ 4.1514692306518555,
189
+ 7.747776031494141,
190
+ -6.167545795440674,
191
+ -5.1719512939453125,
192
+ 0.9890940189361572,
193
+ 4.375288486480713,
194
+ -5.5168561935424805,
195
+ 10.58459758758545
196
+ ],
197
+ "model_value": 0.4061877727508545,
198
+ "model_checkpoint": "r42_1500M.pkl",
199
+ "model_architecture": "simba_aux"
200
+ },
201
+ {
202
+ "turn": 3,
203
+ "player": 0,
204
+ "phase": "draw",
205
+ "action": 0,
206
+ "action_desc": "Human drew from stock",
207
+ "card_drawn": 22,
208
+ "hands": [
209
+ [
210
+ 6,
211
+ 18,
212
+ 44,
213
+ 32,
214
+ 29,
215
+ 51,
216
+ 1,
217
+ 4,
218
+ 12,
219
+ 49
220
+ ],
221
+ [
222
+ 34,
223
+ 17,
224
+ 37,
225
+ 35,
226
+ 47,
227
+ 39,
228
+ 26,
229
+ 50,
230
+ 10,
231
+ 38
232
+ ]
233
+ ],
234
+ "hand_sizes": [
235
+ 10,
236
+ 10
237
+ ],
238
+ "discard_pile": [
239
+ 3
240
+ ],
241
+ "discard_size": 1,
242
+ "stock_remaining": 31,
243
+ "deadwood": [
244
+ 67,
245
+ 75
246
+ ],
247
+ "model_logits": null,
248
+ "model_value": null,
249
+ "model_checkpoint": "r42_1500M.pkl",
250
+ "model_architecture": "simba_aux"
251
+ },
252
+ {
253
+ "turn": 4,
254
+ "player": 0,
255
+ "phase": "discard",
256
+ "action": 12,
257
+ "action_desc": "Human discarded 10\u2665",
258
+ "card_drawn": null,
259
+ "hands": [
260
+ [
261
+ 6,
262
+ 18,
263
+ 44,
264
+ 32,
265
+ 29,
266
+ 51,
267
+ 1,
268
+ 4,
269
+ 12,
270
+ 49,
271
+ 22
272
+ ],
273
+ [
274
+ 34,
275
+ 17,
276
+ 37,
277
+ 35,
278
+ 47,
279
+ 39,
280
+ 26,
281
+ 50,
282
+ 10,
283
+ 38
284
+ ]
285
+ ],
286
+ "hand_sizes": [
287
+ 11,
288
+ 10
289
+ ],
290
+ "discard_pile": [
291
+ 3
292
+ ],
293
+ "discard_size": 1,
294
+ "stock_remaining": 30,
295
+ "deadwood": [
296
+ 77,
297
+ 75
298
+ ],
299
+ "model_logits": null,
300
+ "model_value": null,
301
+ "model_checkpoint": "r42_1500M.pkl",
302
+ "model_architecture": "simba_aux"
303
+ },
304
+ {
305
+ "turn": 5,
306
+ "player": 1,
307
+ "phase": "draw",
308
+ "action": 0,
309
+ "action_desc": "Model drew from stock",
310
+ "card_drawn": 2,
311
+ "hands": [
312
+ [
313
+ 6,
314
+ 18,
315
+ 44,
316
+ 32,
317
+ 29,
318
+ 51,
319
+ 1,
320
+ 4,
321
+ 12,
322
+ 49
323
+ ],
324
+ [
325
+ 34,
326
+ 17,
327
+ 37,
328
+ 35,
329
+ 47,
330
+ 39,
331
+ 26,
332
+ 50,
333
+ 10,
334
+ 38
335
+ ]
336
+ ],
337
+ "hand_sizes": [
338
+ 10,
339
+ 10
340
+ ],
341
+ "discard_pile": [
342
+ 3,
343
+ 22
344
+ ],
345
+ "discard_size": 2,
346
+ "stock_remaining": 30,
347
+ "deadwood": [
348
+ 67,
349
+ 75
350
+ ],
351
+ "model_logits": [
352
+ 1.1383121013641357,
353
+ -1.1441981792449951,
354
+ -5.942113399505615,
355
+ 5.051498889923096,
356
+ -5.376381874084473,
357
+ -5.119042873382568,
358
+ 1.5007623434066772,
359
+ 0.8521668910980225,
360
+ 0.3776625990867615,
361
+ 1.560388207435608,
362
+ -2.3498289585113525,
363
+ 0.042993977665901184,
364
+ 3.9714221954345703,
365
+ 2.4459753036499023,
366
+ -3.931074619293213,
367
+ 13.7251615524292
368
+ ],
369
+ "model_value": 0.2950447201728821,
370
+ "model_checkpoint": "r42_1500M.pkl",
371
+ "model_architecture": "simba_aux"
372
+ },
373
+ {
374
+ "turn": 6,
375
+ "player": 1,
376
+ "phase": "discard",
377
+ "action": 9,
378
+ "action_desc": "Model discarded Q\u2663",
379
+ "card_drawn": null,
380
+ "hands": [
381
+ [
382
+ 6,
383
+ 18,
384
+ 44,
385
+ 32,
386
+ 29,
387
+ 51,
388
+ 1,
389
+ 4,
390
+ 12,
391
+ 49
392
+ ],
393
+ [
394
+ 34,
395
+ 17,
396
+ 37,
397
+ 35,
398
+ 47,
399
+ 39,
400
+ 26,
401
+ 50,
402
+ 10,
403
+ 38,
404
+ 2
405
+ ]
406
+ ],
407
+ "hand_sizes": [
408
+ 10,
409
+ 11
410
+ ],
411
+ "discard_pile": [
412
+ 3,
413
+ 22
414
+ ],
415
+ "discard_size": 2,
416
+ "stock_remaining": 29,
417
+ "deadwood": [
418
+ 67,
419
+ 78
420
+ ],
421
+ "model_logits": [
422
+ -0.6609958410263062,
423
+ 0.6511130332946777,
424
+ -6.807436943054199,
425
+ 4.033630847930908,
426
+ -9.408985137939453,
427
+ -0.8451394438743591,
428
+ 0.16887447237968445,
429
+ -0.4609900414943695,
430
+ -1.300255537033081,
431
+ 2.7978124618530273,
432
+ -2.9067656993865967,
433
+ 1.0653128623962402,
434
+ 5.600430011749268,
435
+ 2.914264678955078,
436
+ -3.8414435386657715,
437
+ 8.631331443786621
438
+ ],
439
+ "model_value": 0.3671707510948181,
440
+ "model_checkpoint": "r42_1500M.pkl",
441
+ "model_architecture": "simba_aux"
442
+ },
443
+ {
444
+ "turn": 7,
445
+ "player": 0,
446
+ "phase": "draw",
447
+ "action": 1,
448
+ "action_desc": "Human drew Q\u2663 from discard",
449
+ "card_drawn": 50,
450
+ "hands": [
451
+ [
452
+ 6,
453
+ 18,
454
+ 44,
455
+ 32,
456
+ 29,
457
+ 51,
458
+ 1,
459
+ 4,
460
+ 12,
461
+ 49
462
+ ],
463
+ [
464
+ 34,
465
+ 17,
466
+ 37,
467
+ 35,
468
+ 47,
469
+ 39,
470
+ 26,
471
+ 2,
472
+ 10,
473
+ 38
474
+ ]
475
+ ],
476
+ "hand_sizes": [
477
+ 10,
478
+ 10
479
+ ],
480
+ "discard_pile": [
481
+ 3,
482
+ 22,
483
+ 50
484
+ ],
485
+ "discard_size": 3,
486
+ "stock_remaining": 29,
487
+ "deadwood": [
488
+ 67,
489
+ 68
490
+ ],
491
+ "model_logits": null,
492
+ "model_value": null,
493
+ "model_checkpoint": "r42_1500M.pkl",
494
+ "model_architecture": "simba_aux"
495
+ },
496
+ {
497
+ "turn": 8,
498
+ "player": 0,
499
+ "phase": "discard",
500
+ "action": 10,
501
+ "action_desc": "Human discarded K\u2660",
502
+ "card_drawn": null,
503
+ "hands": [
504
+ [
505
+ 6,
506
+ 18,
507
+ 44,
508
+ 32,
509
+ 29,
510
+ 51,
511
+ 1,
512
+ 4,
513
+ 12,
514
+ 49,
515
+ 50
516
+ ],
517
+ [
518
+ 34,
519
+ 17,
520
+ 37,
521
+ 35,
522
+ 47,
523
+ 39,
524
+ 26,
525
+ 2,
526
+ 10,
527
+ 38
528
+ ]
529
+ ],
530
+ "hand_sizes": [
531
+ 11,
532
+ 10
533
+ ],
534
+ "discard_pile": [
535
+ 3,
536
+ 22
537
+ ],
538
+ "discard_size": 2,
539
+ "stock_remaining": 29,
540
+ "deadwood": [
541
+ 47,
542
+ 68
543
+ ],
544
+ "model_logits": null,
545
+ "model_value": null,
546
+ "model_checkpoint": "r42_1500M.pkl",
547
+ "model_architecture": "simba_aux"
548
+ },
549
+ {
550
+ "turn": 9,
551
+ "player": 1,
552
+ "phase": "draw",
553
+ "action": 0,
554
+ "action_desc": "Model drew from stock",
555
+ "card_drawn": 27,
556
+ "hands": [
557
+ [
558
+ 6,
559
+ 18,
560
+ 44,
561
+ 32,
562
+ 29,
563
+ 51,
564
+ 1,
565
+ 4,
566
+ 50,
567
+ 49
568
+ ],
569
+ [
570
+ 34,
571
+ 17,
572
+ 37,
573
+ 35,
574
+ 47,
575
+ 39,
576
+ 26,
577
+ 2,
578
+ 10,
579
+ 38
580
+ ]
581
+ ],
582
+ "hand_sizes": [
583
+ 10,
584
+ 10
585
+ ],
586
+ "discard_pile": [
587
+ 3,
588
+ 22,
589
+ 12
590
+ ],
591
+ "discard_size": 3,
592
+ "stock_remaining": 29,
593
+ "deadwood": [
594
+ 37,
595
+ 68
596
+ ],
597
+ "model_logits": [
598
+ 2.366832971572876,
599
+ -2.3701179027557373,
600
+ -5.944068431854248,
601
+ 3.027432918548584,
602
+ -1.0992963314056396,
603
+ -3.5956287384033203,
604
+ -0.822180449962616,
605
+ -0.5052788853645325,
606
+ 1.5524468421936035,
607
+ 1.1444395780563354,
608
+ -1.6324397325515747,
609
+ 1.7227895259857178,
610
+ 1.5738787651062012,
611
+ 2.560776948928833,
612
+ -3.956484794616699,
613
+ 12.925423622131348
614
+ ],
615
+ "model_value": 0.23086480796337128,
616
+ "model_checkpoint": "r42_1500M.pkl",
617
+ "model_architecture": "simba_aux"
618
+ },
619
+ {
620
+ "turn": 10,
621
+ "player": 1,
622
+ "phase": "discard",
623
+ "action": 11,
624
+ "action_desc": "Model discarded K\u2666",
625
+ "card_drawn": null,
626
+ "hands": [
627
+ [
628
+ 6,
629
+ 18,
630
+ 44,
631
+ 32,
632
+ 29,
633
+ 51,
634
+ 1,
635
+ 4,
636
+ 50,
637
+ 49
638
+ ],
639
+ [
640
+ 34,
641
+ 17,
642
+ 37,
643
+ 35,
644
+ 47,
645
+ 39,
646
+ 26,
647
+ 2,
648
+ 10,
649
+ 38,
650
+ 27
651
+ ]
652
+ ],
653
+ "hand_sizes": [
654
+ 10,
655
+ 11
656
+ ],
657
+ "discard_pile": [
658
+ 3,
659
+ 22,
660
+ 12
661
+ ],
662
+ "discard_size": 3,
663
+ "stock_remaining": 28,
664
+ "deadwood": [
665
+ 37,
666
+ 70
667
+ ],
668
+ "model_logits": [
669
+ -3.2162718772888184,
670
+ 3.2084224224090576,
671
+ -10.344769477844238,
672
+ 4.146162509918213,
673
+ -0.6355317831039429,
674
+ -1.9956786632537842,
675
+ -1.1993919610977173,
676
+ 0.5562354326248169,
677
+ -4.796133041381836,
678
+ 4.958113670349121,
679
+ -2.4837467670440674,
680
+ 5.129474639892578,
681
+ -2.688124895095825,
682
+ 1.1886348724365234,
683
+ -1.7478801012039185,
684
+ 6.012758731842041
685
+ ],
686
+ "model_value": 0.2852996289730072,
687
+ "model_checkpoint": "r42_1500M.pkl",
688
+ "model_architecture": "simba_aux"
689
+ },
690
+ {
691
+ "turn": 11,
692
+ "player": 0,
693
+ "phase": "draw",
694
+ "action": 0,
695
+ "action_desc": "Human drew from stock",
696
+ "card_drawn": 30,
697
+ "hands": [
698
+ [
699
+ 6,
700
+ 18,
701
+ 44,
702
+ 32,
703
+ 29,
704
+ 51,
705
+ 1,
706
+ 4,
707
+ 50,
708
+ 49
709
+ ],
710
+ [
711
+ 34,
712
+ 17,
713
+ 37,
714
+ 35,
715
+ 47,
716
+ 39,
717
+ 26,
718
+ 2,
719
+ 10,
720
+ 27
721
+ ]
722
+ ],
723
+ "hand_sizes": [
724
+ 10,
725
+ 10
726
+ ],
727
+ "discard_pile": [
728
+ 3,
729
+ 22,
730
+ 12,
731
+ 38
732
+ ],
733
+ "discard_size": 4,
734
+ "stock_remaining": 28,
735
+ "deadwood": [
736
+ 37,
737
+ 60
738
+ ],
739
+ "model_logits": null,
740
+ "model_value": null,
741
+ "model_checkpoint": "r42_1500M.pkl",
742
+ "model_architecture": "simba_aux"
743
+ },
744
+ {
745
+ "turn": 12,
746
+ "player": 0,
747
+ "phase": "discard",
748
+ "action": 5,
749
+ "action_desc": "Human discarded 7\u2666",
750
+ "card_drawn": null,
751
+ "hands": [
752
+ [
753
+ 6,
754
+ 18,
755
+ 44,
756
+ 32,
757
+ 29,
758
+ 51,
759
+ 1,
760
+ 4,
761
+ 50,
762
+ 49,
763
+ 30
764
+ ],
765
+ [
766
+ 34,
767
+ 17,
768
+ 37,
769
+ 35,
770
+ 47,
771
+ 39,
772
+ 26,
773
+ 2,
774
+ 10,
775
+ 27
776
+ ]
777
+ ],
778
+ "hand_sizes": [
779
+ 11,
780
+ 10
781
+ ],
782
+ "discard_pile": [
783
+ 3,
784
+ 22,
785
+ 12,
786
+ 38
787
+ ],
788
+ "discard_size": 4,
789
+ "stock_remaining": 27,
790
+ "deadwood": [
791
+ 42,
792
+ 60
793
+ ],
794
+ "model_logits": null,
795
+ "model_value": null,
796
+ "model_checkpoint": "r42_1500M.pkl",
797
+ "model_architecture": "simba_aux"
798
+ },
799
+ {
800
+ "turn": 13,
801
+ "player": 1,
802
+ "phase": "draw",
803
+ "action": 0,
804
+ "action_desc": "Model drew from stock",
805
+ "card_drawn": 24,
806
+ "hands": [
807
+ [
808
+ 6,
809
+ 18,
810
+ 44,
811
+ 30,
812
+ 29,
813
+ 51,
814
+ 1,
815
+ 4,
816
+ 50,
817
+ 49
818
+ ],
819
+ [
820
+ 34,
821
+ 17,
822
+ 37,
823
+ 35,
824
+ 47,
825
+ 39,
826
+ 26,
827
+ 2,
828
+ 10,
829
+ 27
830
+ ]
831
+ ],
832
+ "hand_sizes": [
833
+ 10,
834
+ 10
835
+ ],
836
+ "discard_pile": [
837
+ 3,
838
+ 22,
839
+ 12,
840
+ 38,
841
+ 32
842
+ ],
843
+ "discard_size": 5,
844
+ "stock_remaining": 27,
845
+ "deadwood": [
846
+ 35,
847
+ 60
848
+ ],
849
+ "model_logits": [
850
+ 7.045526504516602,
851
+ -7.042075157165527,
852
+ -4.243106842041016,
853
+ 1.6327049732208252,
854
+ 4.720146179199219,
855
+ 2.4579925537109375,
856
+ -0.02795782871544361,
857
+ -3.245696783065796,
858
+ -1.138649344444275,
859
+ 1.8017499446868896,
860
+ 3.7995553016662598,
861
+ -1.3267818689346313,
862
+ -3.9924747943878174,
863
+ 3.9127416610717773,
864
+ -4.958354949951172,
865
+ 9.3467435836792
866
+ ],
867
+ "model_value": 0.02323899231851101,
868
+ "model_checkpoint": "r42_1500M.pkl",
869
+ "model_architecture": "simba_aux"
870
+ },
871
+ {
872
+ "turn": 14,
873
+ "player": 1,
874
+ "phase": "discard",
875
+ "action": 3,
876
+ "action_desc": "Model discarded 5\u2665",
877
+ "card_drawn": null,
878
+ "hands": [
879
+ [
880
+ 6,
881
+ 18,
882
+ 44,
883
+ 30,
884
+ 29,
885
+ 51,
886
+ 1,
887
+ 4,
888
+ 50,
889
+ 49
890
+ ],
891
+ [
892
+ 34,
893
+ 17,
894
+ 37,
895
+ 35,
896
+ 47,
897
+ 39,
898
+ 26,
899
+ 2,
900
+ 10,
901
+ 27,
902
+ 24
903
+ ]
904
+ ],
905
+ "hand_sizes": [
906
+ 10,
907
+ 11
908
+ ],
909
+ "discard_pile": [
910
+ 3,
911
+ 22,
912
+ 12,
913
+ 38,
914
+ 32
915
+ ],
916
+ "discard_size": 5,
917
+ "stock_remaining": 26,
918
+ "deadwood": [
919
+ 35,
920
+ 70
921
+ ],
922
+ "model_logits": [
923
+ -8.59103012084961,
924
+ 8.606566429138184,
925
+ -6.260020732879639,
926
+ 3.660506010055542,
927
+ 0.45137104392051697,
928
+ 3.4923276901245117,
929
+ 0.3168691396713257,
930
+ -1.2298228740692139,
931
+ -4.268662452697754,
932
+ 0.08164475113153458,
933
+ 4.302900314331055,
934
+ -6.573135852813721,
935
+ -5.888349533081055,
936
+ 6.557753086090088,
937
+ -6.475064277648926,
938
+ 1.7604516744613647
939
+ ],
940
+ "model_value": 0.1874338835477829,
941
+ "model_checkpoint": "r42_1500M.pkl",
942
+ "model_architecture": "simba_aux"
943
+ },
944
+ {
945
+ "turn": 15,
946
+ "player": 0,
947
+ "phase": "draw",
948
+ "action": 1,
949
+ "action_desc": "Human drew 5\u2665 from discard",
950
+ "card_drawn": 17,
951
+ "hands": [
952
+ [
953
+ 6,
954
+ 18,
955
+ 44,
956
+ 30,
957
+ 29,
958
+ 51,
959
+ 1,
960
+ 4,
961
+ 50,
962
+ 49
963
+ ],
964
+ [
965
+ 34,
966
+ 24,
967
+ 37,
968
+ 35,
969
+ 47,
970
+ 39,
971
+ 26,
972
+ 2,
973
+ 10,
974
+ 27
975
+ ]
976
+ ],
977
+ "hand_sizes": [
978
+ 10,
979
+ 10
980
+ ],
981
+ "discard_pile": [
982
+ 3,
983
+ 22,
984
+ 12,
985
+ 38,
986
+ 32,
987
+ 17
988
+ ],
989
+ "discard_size": 6,
990
+ "stock_remaining": 26,
991
+ "deadwood": [
992
+ 35,
993
+ 65
994
+ ],
995
+ "model_logits": null,
996
+ "model_value": null,
997
+ "model_checkpoint": "r42_1500M.pkl",
998
+ "model_architecture": "simba_aux"
999
+ },
1000
+ {
1001
+ "turn": 16,
1002
+ "player": 0,
1003
+ "phase": "discard",
1004
+ "action": 2,
1005
+ "action_desc": "Human discarded 7\u2660",
1006
+ "card_drawn": null,
1007
+ "hands": [
1008
+ [
1009
+ 6,
1010
+ 18,
1011
+ 44,
1012
+ 30,
1013
+ 29,
1014
+ 51,
1015
+ 1,
1016
+ 4,
1017
+ 50,
1018
+ 49,
1019
+ 17
1020
+ ],
1021
+ [
1022
+ 34,
1023
+ 24,
1024
+ 37,
1025
+ 35,
1026
+ 47,
1027
+ 39,
1028
+ 26,
1029
+ 2,
1030
+ 10,
1031
+ 27
1032
+ ]
1033
+ ],
1034
+ "hand_sizes": [
1035
+ 11,
1036
+ 10
1037
+ ],
1038
+ "discard_pile": [
1039
+ 3,
1040
+ 22,
1041
+ 12,
1042
+ 38,
1043
+ 32
1044
+ ],
1045
+ "discard_size": 5,
1046
+ "stock_remaining": 26,
1047
+ "deadwood": [
1048
+ 25,
1049
+ 65
1050
+ ],
1051
+ "model_logits": null,
1052
+ "model_value": null,
1053
+ "model_checkpoint": "r42_1500M.pkl",
1054
+ "model_architecture": "simba_aux"
1055
+ },
1056
+ {
1057
+ "turn": 17,
1058
+ "player": 1,
1059
+ "phase": "draw",
1060
+ "action": 0,
1061
+ "action_desc": "Model drew from stock",
1062
+ "card_drawn": 41,
1063
+ "hands": [
1064
+ [
1065
+ 17,
1066
+ 18,
1067
+ 44,
1068
+ 30,
1069
+ 29,
1070
+ 51,
1071
+ 1,
1072
+ 4,
1073
+ 50,
1074
+ 49
1075
+ ],
1076
+ [
1077
+ 34,
1078
+ 24,
1079
+ 37,
1080
+ 35,
1081
+ 47,
1082
+ 39,
1083
+ 26,
1084
+ 2,
1085
+ 10,
1086
+ 27
1087
+ ]
1088
+ ],
1089
+ "hand_sizes": [
1090
+ 10,
1091
+ 10
1092
+ ],
1093
+ "discard_pile": [
1094
+ 3,
1095
+ 22,
1096
+ 12,
1097
+ 38,
1098
+ 32,
1099
+ 6
1100
+ ],
1101
+ "discard_size": 6,
1102
+ "stock_remaining": 26,
1103
+ "deadwood": [
1104
+ 18,
1105
+ 65
1106
+ ],
1107
+ "model_logits": [
1108
+ 10.391035079956055,
1109
+ -10.385504722595215,
1110
+ -3.594442844390869,
1111
+ 2.1573967933654785,
1112
+ 1.796376347541809,
1113
+ 4.022472381591797,
1114
+ 2.581428050994873,
1115
+ -1.7003495693206787,
1116
+ -2.339104413986206,
1117
+ -1.227459192276001,
1118
+ 5.1446709632873535,
1119
+ -0.867095947265625,
1120
+ -3.405510187149048,
1121
+ 2.596508741378784,
1122
+ -3.7273216247558594,
1123
+ 9.730903625488281
1124
+ ],
1125
+ "model_value": -0.007593885064125061,
1126
+ "model_checkpoint": "r42_1500M.pkl",
1127
+ "model_architecture": "simba_aux"
1128
+ },
1129
+ {
1130
+ "turn": 18,
1131
+ "player": 1,
1132
+ "phase": "discard",
1133
+ "action": 6,
1134
+ "action_desc": "Model discarded 9\u2663",
1135
+ "card_drawn": null,
1136
+ "hands": [
1137
+ [
1138
+ 17,
1139
+ 18,
1140
+ 44,
1141
+ 30,
1142
+ 29,
1143
+ 51,
1144
+ 1,
1145
+ 4,
1146
+ 50,
1147
+ 49
1148
+ ],
1149
+ [
1150
+ 34,
1151
+ 24,
1152
+ 37,
1153
+ 35,
1154
+ 47,
1155
+ 39,
1156
+ 26,
1157
+ 2,
1158
+ 10,
1159
+ 27,
1160
+ 41
1161
+ ]
1162
+ ],
1163
+ "hand_sizes": [
1164
+ 10,
1165
+ 11
1166
+ ],
1167
+ "discard_pile": [
1168
+ 3,
1169
+ 22,
1170
+ 12,
1171
+ 38,
1172
+ 32,
1173
+ 6
1174
+ ],
1175
+ "discard_size": 6,
1176
+ "stock_remaining": 25,
1177
+ "deadwood": [
1178
+ 18,
1179
+ 68
1180
+ ],
1181
+ "model_logits": [
1182
+ -1.9585539102554321,
1183
+ 1.97481369972229,
1184
+ -6.673481464385986,
1185
+ 4.547785758972168,
1186
+ 3.7683026790618896,
1187
+ 7.043979644775391,
1188
+ 5.403914451599121,
1189
+ -2.0580615997314453,
1190
+ -6.997781276702881,
1191
+ -2.5418920516967773,
1192
+ 6.457330703735352,
1193
+ -5.842326641082764,
1194
+ -10.477293014526367,
1195
+ 3.9206762313842773,
1196
+ -3.7701938152313232,
1197
+ 0.1457006335258484
1198
+ ],
1199
+ "model_value": 0.011251948773860931,
1200
+ "model_checkpoint": "r42_1500M.pkl",
1201
+ "model_architecture": "simba_aux"
1202
+ },
1203
+ {
1204
+ "turn": 19,
1205
+ "player": 0,
1206
+ "phase": "draw",
1207
+ "action": 0,
1208
+ "action_desc": "Human drew from stock",
1209
+ "card_drawn": 45,
1210
+ "hands": [
1211
+ [
1212
+ 17,
1213
+ 18,
1214
+ 44,
1215
+ 30,
1216
+ 29,
1217
+ 51,
1218
+ 1,
1219
+ 4,
1220
+ 50,
1221
+ 49
1222
+ ],
1223
+ [
1224
+ 34,
1225
+ 24,
1226
+ 37,
1227
+ 35,
1228
+ 41,
1229
+ 39,
1230
+ 26,
1231
+ 2,
1232
+ 10,
1233
+ 27
1234
+ ]
1235
+ ],
1236
+ "hand_sizes": [
1237
+ 10,
1238
+ 10
1239
+ ],
1240
+ "discard_pile": [
1241
+ 3,
1242
+ 22,
1243
+ 12,
1244
+ 38,
1245
+ 32,
1246
+ 6,
1247
+ 47
1248
+ ],
1249
+ "discard_size": 7,
1250
+ "stock_remaining": 25,
1251
+ "deadwood": [
1252
+ 18,
1253
+ 59
1254
+ ],
1255
+ "model_logits": null,
1256
+ "model_value": null,
1257
+ "model_checkpoint": "r42_1500M.pkl",
1258
+ "model_architecture": "simba_aux"
1259
+ },
1260
+ {
1261
+ "turn": 20,
1262
+ "player": 0,
1263
+ "phase": "discard",
1264
+ "action": 12,
1265
+ "action_desc": "Human discarded 7\u2663",
1266
+ "card_drawn": null,
1267
+ "hands": [
1268
+ [
1269
+ 17,
1270
+ 18,
1271
+ 44,
1272
+ 30,
1273
+ 29,
1274
+ 51,
1275
+ 1,
1276
+ 4,
1277
+ 50,
1278
+ 49,
1279
+ 45
1280
+ ],
1281
+ [
1282
+ 34,
1283
+ 24,
1284
+ 37,
1285
+ 35,
1286
+ 41,
1287
+ 39,
1288
+ 26,
1289
+ 2,
1290
+ 10,
1291
+ 27
1292
+ ]
1293
+ ],
1294
+ "hand_sizes": [
1295
+ 11,
1296
+ 10
1297
+ ],
1298
+ "discard_pile": [
1299
+ 3,
1300
+ 22,
1301
+ 12,
1302
+ 38,
1303
+ 32,
1304
+ 6,
1305
+ 47
1306
+ ],
1307
+ "discard_size": 7,
1308
+ "stock_remaining": 24,
1309
+ "deadwood": [
1310
+ 25,
1311
+ 59
1312
+ ],
1313
+ "model_logits": null,
1314
+ "model_value": null,
1315
+ "model_checkpoint": "r42_1500M.pkl",
1316
+ "model_architecture": "simba_aux"
1317
+ },
1318
+ {
1319
+ "turn": 21,
1320
+ "player": 1,
1321
+ "phase": "draw",
1322
+ "action": 0,
1323
+ "action_desc": "Model drew from stock",
1324
+ "card_drawn": 7,
1325
+ "hands": [
1326
+ [
1327
+ 17,
1328
+ 18,
1329
+ 44,
1330
+ 30,
1331
+ 29,
1332
+ 51,
1333
+ 1,
1334
+ 4,
1335
+ 50,
1336
+ 49
1337
+ ],
1338
+ [
1339
+ 34,
1340
+ 24,
1341
+ 37,
1342
+ 35,
1343
+ 41,
1344
+ 39,
1345
+ 26,
1346
+ 2,
1347
+ 10,
1348
+ 27
1349
+ ]
1350
+ ],
1351
+ "hand_sizes": [
1352
+ 10,
1353
+ 10
1354
+ ],
1355
+ "discard_pile": [
1356
+ 3,
1357
+ 22,
1358
+ 12,
1359
+ 38,
1360
+ 32,
1361
+ 6,
1362
+ 47,
1363
+ 45
1364
+ ],
1365
+ "discard_size": 8,
1366
+ "stock_remaining": 24,
1367
+ "deadwood": [
1368
+ 18,
1369
+ 59
1370
+ ],
1371
+ "model_logits": [
1372
+ 10.161972045898438,
1373
+ -10.151374816894531,
1374
+ 2.584629535675049,
1375
+ 1.2898544073104858,
1376
+ 4.28520393371582,
1377
+ 3.5210962295532227,
1378
+ -2.1009316444396973,
1379
+ -3.5999045372009277,
1380
+ -2.6765012741088867,
1381
+ -0.7720077633857727,
1382
+ 5.058908462524414,
1383
+ -1.9633378982543945,
1384
+ -4.077866077423096,
1385
+ 4.24149227142334,
1386
+ -5.360098838806152,
1387
+ 9.156037330627441
1388
+ ],
1389
+ "model_value": -0.007789130322635174,
1390
+ "model_checkpoint": "r42_1500M.pkl",
1391
+ "model_architecture": "simba_aux"
1392
+ },
1393
+ {
1394
+ "turn": 22,
1395
+ "player": 1,
1396
+ "phase": "discard",
1397
+ "action": 10,
1398
+ "action_desc": "Model discarded J\u2660",
1399
+ "card_drawn": null,
1400
+ "hands": [
1401
+ [
1402
+ 17,
1403
+ 18,
1404
+ 44,
1405
+ 30,
1406
+ 29,
1407
+ 51,
1408
+ 1,
1409
+ 4,
1410
+ 50,
1411
+ 49
1412
+ ],
1413
+ [
1414
+ 34,
1415
+ 24,
1416
+ 37,
1417
+ 35,
1418
+ 41,
1419
+ 39,
1420
+ 26,
1421
+ 2,
1422
+ 10,
1423
+ 27,
1424
+ 7
1425
+ ]
1426
+ ],
1427
+ "hand_sizes": [
1428
+ 10,
1429
+ 11
1430
+ ],
1431
+ "discard_pile": [
1432
+ 3,
1433
+ 22,
1434
+ 12,
1435
+ 38,
1436
+ 32,
1437
+ 6,
1438
+ 47,
1439
+ 45
1440
+ ],
1441
+ "discard_size": 8,
1442
+ "stock_remaining": 23,
1443
+ "deadwood": [
1444
+ 18,
1445
+ 67
1446
+ ],
1447
+ "model_logits": [
1448
+ 2.9852776527404785,
1449
+ -2.970914602279663,
1450
+ -2.7061269283294678,
1451
+ 1.5649269819259644,
1452
+ 3.7126429080963135,
1453
+ 3.1658835411071777,
1454
+ -4.321950435638428,
1455
+ -5.137014865875244,
1456
+ -6.562143325805664,
1457
+ -2.7711563110351562,
1458
+ 6.525855541229248,
1459
+ -4.206652641296387,
1460
+ 4.05328893661499,
1461
+ 7.284099578857422,
1462
+ -7.973671913146973,
1463
+ 5.460864543914795
1464
+ ],
1465
+ "model_value": 0.017308970913290977,
1466
+ "model_checkpoint": "r42_1500M.pkl",
1467
+ "model_architecture": "simba_aux"
1468
+ },
1469
+ {
1470
+ "turn": 23,
1471
+ "player": 0,
1472
+ "phase": "draw",
1473
+ "action": 0,
1474
+ "action_desc": "Human drew from stock",
1475
+ "card_drawn": 8,
1476
+ "hands": [
1477
+ [
1478
+ 17,
1479
+ 18,
1480
+ 44,
1481
+ 30,
1482
+ 29,
1483
+ 51,
1484
+ 1,
1485
+ 4,
1486
+ 50,
1487
+ 49
1488
+ ],
1489
+ [
1490
+ 34,
1491
+ 24,
1492
+ 37,
1493
+ 35,
1494
+ 41,
1495
+ 39,
1496
+ 26,
1497
+ 2,
1498
+ 7,
1499
+ 27
1500
+ ]
1501
+ ],
1502
+ "hand_sizes": [
1503
+ 10,
1504
+ 10
1505
+ ],
1506
+ "discard_pile": [
1507
+ 3,
1508
+ 22,
1509
+ 12,
1510
+ 38,
1511
+ 32,
1512
+ 6,
1513
+ 47,
1514
+ 45,
1515
+ 10
1516
+ ],
1517
+ "discard_size": 9,
1518
+ "stock_remaining": 23,
1519
+ "deadwood": [
1520
+ 18,
1521
+ 57
1522
+ ],
1523
+ "model_logits": null,
1524
+ "model_value": null,
1525
+ "model_checkpoint": "r42_1500M.pkl",
1526
+ "model_architecture": "simba_aux"
1527
+ },
1528
+ {
1529
+ "turn": 24,
1530
+ "player": 0,
1531
+ "phase": "discard",
1532
+ "action": 12,
1533
+ "action_desc": "Human discarded 9\u2660",
1534
+ "card_drawn": null,
1535
+ "hands": [
1536
+ [
1537
+ 17,
1538
+ 18,
1539
+ 44,
1540
+ 30,
1541
+ 29,
1542
+ 51,
1543
+ 1,
1544
+ 4,
1545
+ 50,
1546
+ 49,
1547
+ 8
1548
+ ],
1549
+ [
1550
+ 34,
1551
+ 24,
1552
+ 37,
1553
+ 35,
1554
+ 41,
1555
+ 39,
1556
+ 26,
1557
+ 2,
1558
+ 7,
1559
+ 27
1560
+ ]
1561
+ ],
1562
+ "hand_sizes": [
1563
+ 11,
1564
+ 10
1565
+ ],
1566
+ "discard_pile": [
1567
+ 3,
1568
+ 22,
1569
+ 12,
1570
+ 38,
1571
+ 32,
1572
+ 6,
1573
+ 47,
1574
+ 45,
1575
+ 10
1576
+ ],
1577
+ "discard_size": 9,
1578
+ "stock_remaining": 22,
1579
+ "deadwood": [
1580
+ 27,
1581
+ 57
1582
+ ],
1583
+ "model_logits": null,
1584
+ "model_value": null,
1585
+ "model_checkpoint": "r42_1500M.pkl",
1586
+ "model_architecture": "simba_aux"
1587
+ },
1588
+ {
1589
+ "turn": 25,
1590
+ "player": 1,
1591
+ "phase": "draw",
1592
+ "action": 0,
1593
+ "action_desc": "Model drew from stock",
1594
+ "card_drawn": 21,
1595
+ "hands": [
1596
+ [
1597
+ 17,
1598
+ 18,
1599
+ 44,
1600
+ 30,
1601
+ 29,
1602
+ 51,
1603
+ 1,
1604
+ 4,
1605
+ 50,
1606
+ 49
1607
+ ],
1608
+ [
1609
+ 34,
1610
+ 24,
1611
+ 37,
1612
+ 35,
1613
+ 41,
1614
+ 39,
1615
+ 26,
1616
+ 2,
1617
+ 7,
1618
+ 27
1619
+ ]
1620
+ ],
1621
+ "hand_sizes": [
1622
+ 10,
1623
+ 10
1624
+ ],
1625
+ "discard_pile": [
1626
+ 3,
1627
+ 22,
1628
+ 12,
1629
+ 38,
1630
+ 32,
1631
+ 6,
1632
+ 47,
1633
+ 45,
1634
+ 10,
1635
+ 8
1636
+ ],
1637
+ "discard_size": 10,
1638
+ "stock_remaining": 22,
1639
+ "deadwood": [
1640
+ 18,
1641
+ 57
1642
+ ],
1643
+ "model_logits": [
1644
+ 8.411372184753418,
1645
+ -8.40450382232666,
1646
+ 5.19318962097168,
1647
+ 0.9832573533058167,
1648
+ 3.67116379737854,
1649
+ 2.46591854095459,
1650
+ -1.3680436611175537,
1651
+ -4.1311445236206055,
1652
+ -1.016548752784729,
1653
+ -1.8395800590515137,
1654
+ 4.74591588973999,
1655
+ -3.7313156127929688,
1656
+ -4.703254699707031,
1657
+ 3.904726982116699,
1658
+ -5.172491073608398,
1659
+ 10.954968452453613
1660
+ ],
1661
+ "model_value": 0.014860923402011395,
1662
+ "model_checkpoint": "r42_1500M.pkl",
1663
+ "model_architecture": "simba_aux"
1664
+ },
1665
+ {
1666
+ "turn": 26,
1667
+ "player": 1,
1668
+ "phase": "discard",
1669
+ "action": 12,
1670
+ "action_desc": "Model discarded 9\u2665",
1671
+ "card_drawn": null,
1672
+ "hands": [
1673
+ [
1674
+ 17,
1675
+ 18,
1676
+ 44,
1677
+ 30,
1678
+ 29,
1679
+ 51,
1680
+ 1,
1681
+ 4,
1682
+ 50,
1683
+ 49
1684
+ ],
1685
+ [
1686
+ 34,
1687
+ 24,
1688
+ 37,
1689
+ 35,
1690
+ 41,
1691
+ 39,
1692
+ 26,
1693
+ 2,
1694
+ 7,
1695
+ 27,
1696
+ 21
1697
+ ]
1698
+ ],
1699
+ "hand_sizes": [
1700
+ 10,
1701
+ 11
1702
+ ],
1703
+ "discard_pile": [
1704
+ 3,
1705
+ 22,
1706
+ 12,
1707
+ 38,
1708
+ 32,
1709
+ 6,
1710
+ 47,
1711
+ 45,
1712
+ 10,
1713
+ 8
1714
+ ],
1715
+ "discard_size": 10,
1716
+ "stock_remaining": 21,
1717
+ "deadwood": [
1718
+ 18,
1719
+ 66
1720
+ ],
1721
+ "model_logits": [
1722
+ -7.719825744628906,
1723
+ 7.7204999923706055,
1724
+ 3.7984530925750732,
1725
+ 0.5445335507392883,
1726
+ 0.7104087471961975,
1727
+ -0.07493980973958969,
1728
+ -5.383929252624512,
1729
+ -8.737409591674805,
1730
+ -4.760289192199707,
1731
+ -4.334479808807373,
1732
+ 5.58394718170166,
1733
+ -8.905295372009277,
1734
+ 10.232336044311523,
1735
+ 9.107224464416504,
1736
+ -9.769830703735352,
1737
+ 6.859590530395508
1738
+ ],
1739
+ "model_value": 0.1022002249956131,
1740
+ "model_checkpoint": "r42_1500M.pkl",
1741
+ "model_architecture": "simba_aux"
1742
+ },
1743
+ {
1744
+ "turn": 27,
1745
+ "player": 0,
1746
+ "phase": "draw",
1747
+ "action": 0,
1748
+ "action_desc": "Human drew from stock",
1749
+ "card_drawn": 11,
1750
+ "hands": [
1751
+ [
1752
+ 17,
1753
+ 18,
1754
+ 44,
1755
+ 30,
1756
+ 29,
1757
+ 51,
1758
+ 1,
1759
+ 4,
1760
+ 50,
1761
+ 49
1762
+ ],
1763
+ [
1764
+ 34,
1765
+ 24,
1766
+ 37,
1767
+ 35,
1768
+ 41,
1769
+ 39,
1770
+ 26,
1771
+ 2,
1772
+ 7,
1773
+ 27
1774
+ ]
1775
+ ],
1776
+ "hand_sizes": [
1777
+ 10,
1778
+ 10
1779
+ ],
1780
+ "discard_pile": [
1781
+ 3,
1782
+ 22,
1783
+ 12,
1784
+ 38,
1785
+ 32,
1786
+ 6,
1787
+ 47,
1788
+ 45,
1789
+ 10,
1790
+ 8,
1791
+ 21
1792
+ ],
1793
+ "discard_size": 11,
1794
+ "stock_remaining": 21,
1795
+ "deadwood": [
1796
+ 18,
1797
+ 57
1798
+ ],
1799
+ "model_logits": null,
1800
+ "model_value": null,
1801
+ "model_checkpoint": "r42_1500M.pkl",
1802
+ "model_architecture": "simba_aux"
1803
+ },
1804
+ {
1805
+ "turn": 28,
1806
+ "player": 0,
1807
+ "phase": "discard",
1808
+ "action": 12,
1809
+ "action_desc": "Human discarded Q\u2660",
1810
+ "card_drawn": null,
1811
+ "hands": [
1812
+ [
1813
+ 17,
1814
+ 18,
1815
+ 44,
1816
+ 30,
1817
+ 29,
1818
+ 51,
1819
+ 1,
1820
+ 4,
1821
+ 50,
1822
+ 49,
1823
+ 11
1824
+ ],
1825
+ [
1826
+ 34,
1827
+ 24,
1828
+ 37,
1829
+ 35,
1830
+ 41,
1831
+ 39,
1832
+ 26,
1833
+ 2,
1834
+ 7,
1835
+ 27
1836
+ ]
1837
+ ],
1838
+ "hand_sizes": [
1839
+ 11,
1840
+ 10
1841
+ ],
1842
+ "discard_pile": [
1843
+ 3,
1844
+ 22,
1845
+ 12,
1846
+ 38,
1847
+ 32,
1848
+ 6,
1849
+ 47,
1850
+ 45,
1851
+ 10,
1852
+ 8,
1853
+ 21
1854
+ ],
1855
+ "discard_size": 11,
1856
+ "stock_remaining": 20,
1857
+ "deadwood": [
1858
+ 28,
1859
+ 57
1860
+ ],
1861
+ "model_logits": null,
1862
+ "model_value": null,
1863
+ "model_checkpoint": "r42_1500M.pkl",
1864
+ "model_architecture": "simba_aux"
1865
+ },
1866
+ {
1867
+ "turn": 29,
1868
+ "player": 1,
1869
+ "phase": "draw",
1870
+ "action": 1,
1871
+ "action_desc": "Model drew Q\u2660 from discard",
1872
+ "card_drawn": 11,
1873
+ "hands": [
1874
+ [
1875
+ 17,
1876
+ 18,
1877
+ 44,
1878
+ 30,
1879
+ 29,
1880
+ 51,
1881
+ 1,
1882
+ 4,
1883
+ 50,
1884
+ 49
1885
+ ],
1886
+ [
1887
+ 34,
1888
+ 24,
1889
+ 37,
1890
+ 35,
1891
+ 41,
1892
+ 39,
1893
+ 26,
1894
+ 2,
1895
+ 7,
1896
+ 27
1897
+ ]
1898
+ ],
1899
+ "hand_sizes": [
1900
+ 10,
1901
+ 10
1902
+ ],
1903
+ "discard_pile": [
1904
+ 3,
1905
+ 22,
1906
+ 12,
1907
+ 38,
1908
+ 32,
1909
+ 6,
1910
+ 47,
1911
+ 45,
1912
+ 10,
1913
+ 8,
1914
+ 21,
1915
+ 11
1916
+ ],
1917
+ "discard_size": 12,
1918
+ "stock_remaining": 20,
1919
+ "deadwood": [
1920
+ 18,
1921
+ 57
1922
+ ],
1923
+ "model_logits": [
1924
+ -24.93402671813965,
1925
+ 24.942411422729492,
1926
+ 7.378296375274658,
1927
+ 0.7266867160797119,
1928
+ 3.8564817905426025,
1929
+ -2.0465314388275146,
1930
+ -0.979244589805603,
1931
+ -6.419928550720215,
1932
+ -0.935878574848175,
1933
+ -1.548100233078003,
1934
+ 1.4173760414123535,
1935
+ -5.132580757141113,
1936
+ -10.507147789001465,
1937
+ 1.0401138067245483,
1938
+ -0.34054768085479736,
1939
+ -1.8066574335098267
1940
+ ],
1941
+ "model_value": 0.07394938915967941,
1942
+ "model_checkpoint": "r42_1500M.pkl",
1943
+ "model_architecture": "simba_aux"
1944
+ },
1945
+ {
1946
+ "turn": 30,
1947
+ "player": 1,
1948
+ "phase": "discard",
1949
+ "action": 2,
1950
+ "action_desc": "Model discarded 9\u2666",
1951
+ "card_drawn": null,
1952
+ "hands": [
1953
+ [
1954
+ 17,
1955
+ 18,
1956
+ 44,
1957
+ 30,
1958
+ 29,
1959
+ 51,
1960
+ 1,
1961
+ 4,
1962
+ 50,
1963
+ 49
1964
+ ],
1965
+ [
1966
+ 34,
1967
+ 24,
1968
+ 37,
1969
+ 35,
1970
+ 41,
1971
+ 39,
1972
+ 26,
1973
+ 2,
1974
+ 7,
1975
+ 27,
1976
+ 11
1977
+ ]
1978
+ ],
1979
+ "hand_sizes": [
1980
+ 10,
1981
+ 11
1982
+ ],
1983
+ "discard_pile": [
1984
+ 3,
1985
+ 22,
1986
+ 12,
1987
+ 38,
1988
+ 32,
1989
+ 6,
1990
+ 47,
1991
+ 45,
1992
+ 10,
1993
+ 8,
1994
+ 21
1995
+ ],
1996
+ "discard_size": 11,
1997
+ "stock_remaining": 20,
1998
+ "deadwood": [
1999
+ 18,
2000
+ 37
2001
+ ],
2002
+ "model_logits": [
2003
+ -10.94430160522461,
2004
+ 10.94465446472168,
2005
+ 17.222379684448242,
2006
+ -10.62421703338623,
2007
+ -12.059831619262695,
2008
+ 8.573034286499023,
2009
+ -1.2362626791000366,
2010
+ -2.501908779144287,
2011
+ -3.467503786087036,
2012
+ 0.42252662777900696,
2013
+ 9.264952659606934,
2014
+ -2.08377742767334,
2015
+ -12.868328094482422,
2016
+ 0.9883936047554016,
2017
+ -0.5828156471252441,
2018
+ -0.4634231626987457
2019
+ ],
2020
+ "model_value": 0.12962187826633453,
2021
+ "model_checkpoint": "r42_1500M.pkl",
2022
+ "model_architecture": "simba_aux"
2023
+ },
2024
+ {
2025
+ "turn": 31,
2026
+ "player": 0,
2027
+ "phase": "draw",
2028
+ "action": 0,
2029
+ "action_desc": "Human drew from stock",
2030
+ "card_drawn": 23,
2031
+ "hands": [
2032
+ [
2033
+ 17,
2034
+ 18,
2035
+ 44,
2036
+ 30,
2037
+ 29,
2038
+ 51,
2039
+ 1,
2040
+ 4,
2041
+ 50,
2042
+ 49
2043
+ ],
2044
+ [
2045
+ 11,
2046
+ 24,
2047
+ 37,
2048
+ 35,
2049
+ 41,
2050
+ 39,
2051
+ 26,
2052
+ 2,
2053
+ 7,
2054
+ 27
2055
+ ]
2056
+ ],
2057
+ "hand_sizes": [
2058
+ 10,
2059
+ 10
2060
+ ],
2061
+ "discard_pile": [
2062
+ 3,
2063
+ 22,
2064
+ 12,
2065
+ 38,
2066
+ 32,
2067
+ 6,
2068
+ 47,
2069
+ 45,
2070
+ 10,
2071
+ 8,
2072
+ 21,
2073
+ 34
2074
+ ],
2075
+ "discard_size": 12,
2076
+ "stock_remaining": 20,
2077
+ "deadwood": [
2078
+ 18,
2079
+ 28
2080
+ ],
2081
+ "model_logits": null,
2082
+ "model_value": null,
2083
+ "model_checkpoint": "r42_1500M.pkl",
2084
+ "model_architecture": "simba_aux"
2085
+ },
2086
+ {
2087
+ "turn": 32,
2088
+ "player": 0,
2089
+ "phase": "discard",
2090
+ "action": 12,
2091
+ "action_desc": "Human discarded J\u2665",
2092
+ "card_drawn": null,
2093
+ "hands": [
2094
+ [
2095
+ 17,
2096
+ 18,
2097
+ 44,
2098
+ 30,
2099
+ 29,
2100
+ 51,
2101
+ 1,
2102
+ 4,
2103
+ 50,
2104
+ 49,
2105
+ 23
2106
+ ],
2107
+ [
2108
+ 11,
2109
+ 24,
2110
+ 37,
2111
+ 35,
2112
+ 41,
2113
+ 39,
2114
+ 26,
2115
+ 2,
2116
+ 7,
2117
+ 27
2118
+ ]
2119
+ ],
2120
+ "hand_sizes": [
2121
+ 11,
2122
+ 10
2123
+ ],
2124
+ "discard_pile": [
2125
+ 3,
2126
+ 22,
2127
+ 12,
2128
+ 38,
2129
+ 32,
2130
+ 6,
2131
+ 47,
2132
+ 45,
2133
+ 10,
2134
+ 8,
2135
+ 21,
2136
+ 34
2137
+ ],
2138
+ "discard_size": 12,
2139
+ "stock_remaining": 19,
2140
+ "deadwood": [
2141
+ 28,
2142
+ 28
2143
+ ],
2144
+ "model_logits": null,
2145
+ "model_value": null,
2146
+ "model_checkpoint": "r42_1500M.pkl",
2147
+ "model_architecture": "simba_aux"
2148
+ },
2149
+ {
2150
+ "turn": 33,
2151
+ "player": 1,
2152
+ "phase": "draw",
2153
+ "action": 0,
2154
+ "action_desc": "Model drew from stock",
2155
+ "card_drawn": 5,
2156
+ "hands": [
2157
+ [
2158
+ 17,
2159
+ 18,
2160
+ 44,
2161
+ 30,
2162
+ 29,
2163
+ 51,
2164
+ 1,
2165
+ 4,
2166
+ 50,
2167
+ 49
2168
+ ],
2169
+ [
2170
+ 11,
2171
+ 24,
2172
+ 37,
2173
+ 35,
2174
+ 41,
2175
+ 39,
2176
+ 26,
2177
+ 2,
2178
+ 7,
2179
+ 27
2180
+ ]
2181
+ ],
2182
+ "hand_sizes": [
2183
+ 10,
2184
+ 10
2185
+ ],
2186
+ "discard_pile": [
2187
+ 3,
2188
+ 22,
2189
+ 12,
2190
+ 38,
2191
+ 32,
2192
+ 6,
2193
+ 47,
2194
+ 45,
2195
+ 10,
2196
+ 8,
2197
+ 21,
2198
+ 34,
2199
+ 23
2200
+ ],
2201
+ "discard_size": 13,
2202
+ "stock_remaining": 19,
2203
+ "deadwood": [
2204
+ 18,
2205
+ 28
2206
+ ],
2207
+ "model_logits": [
2208
+ 9.565999031066895,
2209
+ -9.580016136169434,
2210
+ -7.211187839508057,
2211
+ -7.344688892364502,
2212
+ -10.900849342346191,
2213
+ 14.601959228515625,
2214
+ 1.9691658020019531,
2215
+ -0.9385589361190796,
2216
+ 1.1075159311294556,
2217
+ 0.24079486727714539,
2218
+ 9.675326347351074,
2219
+ 0.0012815757654607296,
2220
+ -0.666558027267456,
2221
+ -0.4669443666934967,
2222
+ -0.649480938911438,
2223
+ 9.554314613342285
2224
+ ],
2225
+ "model_value": 0.12151311337947845,
2226
+ "model_checkpoint": "r42_1500M.pkl",
2227
+ "model_architecture": "simba_aux"
2228
+ },
2229
+ {
2230
+ "turn": 34,
2231
+ "player": 1,
2232
+ "phase": "discard",
2233
+ "action": 5,
2234
+ "action_desc": "Model discarded 10\u2666",
2235
+ "card_drawn": null,
2236
+ "hands": [
2237
+ [
2238
+ 17,
2239
+ 18,
2240
+ 44,
2241
+ 30,
2242
+ 29,
2243
+ 51,
2244
+ 1,
2245
+ 4,
2246
+ 50,
2247
+ 49
2248
+ ],
2249
+ [
2250
+ 11,
2251
+ 24,
2252
+ 37,
2253
+ 35,
2254
+ 41,
2255
+ 39,
2256
+ 26,
2257
+ 2,
2258
+ 7,
2259
+ 27,
2260
+ 5
2261
+ ]
2262
+ ],
2263
+ "hand_sizes": [
2264
+ 10,
2265
+ 11
2266
+ ],
2267
+ "discard_pile": [
2268
+ 3,
2269
+ 22,
2270
+ 12,
2271
+ 38,
2272
+ 32,
2273
+ 6,
2274
+ 47,
2275
+ 45,
2276
+ 10,
2277
+ 8,
2278
+ 21,
2279
+ 34,
2280
+ 23
2281
+ ],
2282
+ "discard_size": 13,
2283
+ "stock_remaining": 18,
2284
+ "deadwood": [
2285
+ 18,
2286
+ 34
2287
+ ],
2288
+ "model_logits": [
2289
+ 3.215306043624878,
2290
+ -3.222439765930176,
2291
+ -13.564806938171387,
2292
+ -12.08281421661377,
2293
+ -13.280232429504395,
2294
+ 17.127361297607422,
2295
+ 2.2496299743652344,
2296
+ -1.8122936487197876,
2297
+ -1.1188228130340576,
2298
+ -0.20448243618011475,
2299
+ 9.995448112487793,
2300
+ -2.3092570304870605,
2301
+ 8.856048583984375,
2302
+ 2.5617918968200684,
2303
+ -3.259829044342041,
2304
+ 6.011346340179443
2305
+ ],
2306
+ "model_value": 0.13926514983177185,
2307
+ "model_checkpoint": "r42_1500M.pkl",
2308
+ "model_architecture": "simba_aux"
2309
+ },
2310
+ {
2311
+ "turn": 35,
2312
+ "player": 0,
2313
+ "phase": "draw",
2314
+ "action": 0,
2315
+ "action_desc": "Human drew from stock",
2316
+ "card_drawn": 15,
2317
+ "hands": [
2318
+ [
2319
+ 17,
2320
+ 18,
2321
+ 44,
2322
+ 30,
2323
+ 29,
2324
+ 51,
2325
+ 1,
2326
+ 4,
2327
+ 50,
2328
+ 49
2329
+ ],
2330
+ [
2331
+ 11,
2332
+ 24,
2333
+ 37,
2334
+ 5,
2335
+ 41,
2336
+ 39,
2337
+ 26,
2338
+ 2,
2339
+ 7,
2340
+ 27
2341
+ ]
2342
+ ],
2343
+ "hand_sizes": [
2344
+ 10,
2345
+ 10
2346
+ ],
2347
+ "discard_pile": [
2348
+ 3,
2349
+ 22,
2350
+ 12,
2351
+ 38,
2352
+ 32,
2353
+ 6,
2354
+ 47,
2355
+ 45,
2356
+ 10,
2357
+ 8,
2358
+ 21,
2359
+ 34,
2360
+ 23,
2361
+ 35
2362
+ ],
2363
+ "discard_size": 14,
2364
+ "stock_remaining": 18,
2365
+ "deadwood": [
2366
+ 18,
2367
+ 24
2368
+ ],
2369
+ "model_logits": null,
2370
+ "model_value": null,
2371
+ "model_checkpoint": "r42_1500M.pkl",
2372
+ "model_architecture": "simba_aux"
2373
+ },
2374
+ {
2375
+ "turn": 36,
2376
+ "player": 0,
2377
+ "phase": "discard",
2378
+ "action": 12,
2379
+ "action_desc": "Human discarded 3\u2665",
2380
+ "card_drawn": null,
2381
+ "hands": [
2382
+ [
2383
+ 17,
2384
+ 18,
2385
+ 44,
2386
+ 30,
2387
+ 29,
2388
+ 51,
2389
+ 1,
2390
+ 4,
2391
+ 50,
2392
+ 49,
2393
+ 15
2394
+ ],
2395
+ [
2396
+ 11,
2397
+ 24,
2398
+ 37,
2399
+ 5,
2400
+ 41,
2401
+ 39,
2402
+ 26,
2403
+ 2,
2404
+ 7,
2405
+ 27
2406
+ ]
2407
+ ],
2408
+ "hand_sizes": [
2409
+ 11,
2410
+ 10
2411
+ ],
2412
+ "discard_pile": [
2413
+ 3,
2414
+ 22,
2415
+ 12,
2416
+ 38,
2417
+ 32,
2418
+ 6,
2419
+ 47,
2420
+ 45,
2421
+ 10,
2422
+ 8,
2423
+ 21,
2424
+ 34,
2425
+ 23,
2426
+ 35
2427
+ ],
2428
+ "discard_size": 14,
2429
+ "stock_remaining": 17,
2430
+ "deadwood": [
2431
+ 21,
2432
+ 24
2433
+ ],
2434
+ "model_logits": null,
2435
+ "model_value": null,
2436
+ "model_checkpoint": "r42_1500M.pkl",
2437
+ "model_architecture": "simba_aux"
2438
+ },
2439
+ {
2440
+ "turn": 37,
2441
+ "player": 1,
2442
+ "phase": "draw",
2443
+ "action": 1,
2444
+ "action_desc": "Model drew 3\u2665 from discard",
2445
+ "card_drawn": 15,
2446
+ "hands": [
2447
+ [
2448
+ 17,
2449
+ 18,
2450
+ 44,
2451
+ 30,
2452
+ 29,
2453
+ 51,
2454
+ 1,
2455
+ 4,
2456
+ 50,
2457
+ 49
2458
+ ],
2459
+ [
2460
+ 11,
2461
+ 24,
2462
+ 37,
2463
+ 5,
2464
+ 41,
2465
+ 39,
2466
+ 26,
2467
+ 2,
2468
+ 7,
2469
+ 27
2470
+ ]
2471
+ ],
2472
+ "hand_sizes": [
2473
+ 10,
2474
+ 10
2475
+ ],
2476
+ "discard_pile": [
2477
+ 3,
2478
+ 22,
2479
+ 12,
2480
+ 38,
2481
+ 32,
2482
+ 6,
2483
+ 47,
2484
+ 45,
2485
+ 10,
2486
+ 8,
2487
+ 21,
2488
+ 34,
2489
+ 23,
2490
+ 35,
2491
+ 15
2492
+ ],
2493
+ "discard_size": 15,
2494
+ "stock_remaining": 17,
2495
+ "deadwood": [
2496
+ 18,
2497
+ 24
2498
+ ],
2499
+ "model_logits": [
2500
+ -12.232564926147461,
2501
+ 12.219988822937012,
2502
+ -10.495654106140137,
2503
+ -9.973735809326172,
2504
+ -12.007347106933594,
2505
+ 7.85431432723999,
2506
+ 8.267888069152832,
2507
+ 1.1165764331817627,
2508
+ 0.21806837618350983,
2509
+ 4.128976345062256,
2510
+ 9.987337112426758,
2511
+ -1.035211205482483,
2512
+ -6.678713798522949,
2513
+ -3.1516964435577393,
2514
+ 3.233336925506592,
2515
+ 2.4673125743865967
2516
+ ],
2517
+ "model_value": 0.09847187995910645,
2518
+ "model_checkpoint": "r42_1500M.pkl",
2519
+ "model_architecture": "simba_aux"
2520
+ },
2521
+ {
2522
+ "turn": 38,
2523
+ "player": 1,
2524
+ "phase": "discard",
2525
+ "action": 10,
2526
+ "action_desc": "Model discarded 8\u2660",
2527
+ "card_drawn": null,
2528
+ "hands": [
2529
+ [
2530
+ 17,
2531
+ 18,
2532
+ 44,
2533
+ 30,
2534
+ 29,
2535
+ 51,
2536
+ 1,
2537
+ 4,
2538
+ 50,
2539
+ 49
2540
+ ],
2541
+ [
2542
+ 11,
2543
+ 24,
2544
+ 37,
2545
+ 5,
2546
+ 41,
2547
+ 39,
2548
+ 26,
2549
+ 2,
2550
+ 7,
2551
+ 27,
2552
+ 15
2553
+ ]
2554
+ ],
2555
+ "hand_sizes": [
2556
+ 10,
2557
+ 11
2558
+ ],
2559
+ "discard_pile": [
2560
+ 3,
2561
+ 22,
2562
+ 12,
2563
+ 38,
2564
+ 32,
2565
+ 6,
2566
+ 47,
2567
+ 45,
2568
+ 10,
2569
+ 8,
2570
+ 21,
2571
+ 34,
2572
+ 23,
2573
+ 35
2574
+ ],
2575
+ "discard_size": 14,
2576
+ "stock_remaining": 17,
2577
+ "deadwood": [
2578
+ 18,
2579
+ 18
2580
+ ],
2581
+ "model_logits": [
2582
+ 1.4365679025650024,
2583
+ -1.446953535079956,
2584
+ -16.10041618347168,
2585
+ -15.109973907470703,
2586
+ -14.07232666015625,
2587
+ 16.227991104125977,
2588
+ 0.5810877680778503,
2589
+ 3.108476161956787,
2590
+ 3.7765657901763916,
2591
+ -2.12117600440979,
2592
+ 16.330108642578125,
2593
+ 4.332938194274902,
2594
+ -2.202576160430908,
2595
+ -0.3592878580093384,
2596
+ 0.22968317568302155,
2597
+ 1.6667919158935547
2598
+ ],
2599
+ "model_value": 0.31257057189941406,
2600
+ "model_checkpoint": "r42_1500M.pkl",
2601
+ "model_architecture": "simba_aux"
2602
+ },
2603
+ {
2604
+ "turn": 39,
2605
+ "player": 1,
2606
+ "phase": "knock_decision",
2607
+ "action": 14,
2608
+ "action_desc": "Model knocked",
2609
+ "card_drawn": null,
2610
+ "hands": [
2611
+ [
2612
+ 17,
2613
+ 18,
2614
+ 44,
2615
+ 30,
2616
+ 29,
2617
+ 51,
2618
+ 1,
2619
+ 4,
2620
+ 50,
2621
+ 49
2622
+ ],
2623
+ [
2624
+ 11,
2625
+ 24,
2626
+ 37,
2627
+ 5,
2628
+ 41,
2629
+ 39,
2630
+ 26,
2631
+ 2,
2632
+ 15,
2633
+ 27
2634
+ ]
2635
+ ],
2636
+ "hand_sizes": [
2637
+ 10,
2638
+ 10
2639
+ ],
2640
+ "discard_pile": [
2641
+ 3,
2642
+ 22,
2643
+ 12,
2644
+ 38,
2645
+ 32,
2646
+ 6,
2647
+ 47,
2648
+ 45,
2649
+ 10,
2650
+ 8,
2651
+ 21,
2652
+ 34,
2653
+ 23,
2654
+ 35,
2655
+ 7
2656
+ ],
2657
+ "discard_size": 15,
2658
+ "stock_remaining": 17,
2659
+ "deadwood": [
2660
+ 18,
2661
+ 10
2662
+ ],
2663
+ "model_logits": [
2664
+ 5.006402015686035,
2665
+ -5.0432024002075195,
2666
+ -13.783223152160645,
2667
+ -9.60841178894043,
2668
+ -10.411337852478027,
2669
+ 11.436795234680176,
2670
+ 3.947007179260254,
2671
+ 0.7098307609558105,
2672
+ 5.296677589416504,
2673
+ 2.4295613765716553,
2674
+ -1.9125772714614868,
2675
+ 4.476025581359863,
2676
+ 5.673623561859131,
2677
+ -5.835636615753174,
2678
+ 4.4886603355407715,
2679
+ 11.009228706359863
2680
+ ],
2681
+ "model_value": 0.27584129571914673,
2682
+ "model_checkpoint": "r42_1500M.pkl",
2683
+ "model_architecture": "simba_aux"
2684
+ },
2685
+ {
2686
+ "turn": 40,
2687
+ "player": -1,
2688
+ "phase": "game_over",
2689
+ "action": -1,
2690
+ "action_desc": "Game over.",
2691
+ "card_drawn": null,
2692
+ "hands": [
2693
+ [
2694
+ 17,
2695
+ 18,
2696
+ 44,
2697
+ 30,
2698
+ 29,
2699
+ 51,
2700
+ 1,
2701
+ 4,
2702
+ 50,
2703
+ 49
2704
+ ],
2705
+ [
2706
+ 11,
2707
+ 24,
2708
+ 37,
2709
+ 5,
2710
+ 41,
2711
+ 39,
2712
+ 26,
2713
+ 2,
2714
+ 15,
2715
+ 27
2716
+ ]
2717
+ ],
2718
+ "hand_sizes": [
2719
+ 10,
2720
+ 10
2721
+ ],
2722
+ "discard_pile": [
2723
+ 3,
2724
+ 22,
2725
+ 12,
2726
+ 38,
2727
+ 32,
2728
+ 6,
2729
+ 47,
2730
+ 45,
2731
+ 10,
2732
+ 8,
2733
+ 21,
2734
+ 34,
2735
+ 23,
2736
+ 35,
2737
+ 7
2738
+ ],
2739
+ "discard_size": 15,
2740
+ "stock_remaining": 17,
2741
+ "deadwood": [
2742
+ 18,
2743
+ 10
2744
+ ],
2745
+ "model_logits": null,
2746
+ "model_value": null,
2747
+ "model_checkpoint": "r42_1500M.pkl",
2748
+ "model_architecture": "simba_aux"
2749
+ }
2750
+ ]
2751
+ }