description stringlengths 9 136 | template_id stringclasses 55
values | template_parameters stringlengths 2 132 | component_families stringclasses 17
values | split_type stringclasses 2
values | semantic_features stringclasses 144
values | history_depth int64 0 16 | period_or_chunk_length int64 0 12 | branch_count int64 0 4 | attribute_cardinality int64 0 24 | ast_nodes int64 13 91 | acceptance_rate float64 0.08 0.92 | acceptance_entropy float64 0.39 1 | signature_hash stringlengths 64 64 | designed_level int64 0 10 | frontier_solve_rate float64 -1 -1 | mean_reward float64 -1 -1 | difficulty_band stringclasses 1
value | calibration_model_panel stringclasses 1
value | dataset_version stringclasses 1
value | rule_id stringlengths 14 70 | label stringlengths 9 136 | family stringclasses 13
values | code stringlengths 21 256 | protocol_version stringclasses 1
value | calibration_objective stringclasses 1
value | calibration_horizon int64 100 100 | legacy_horizon int64 30 30 | complexity_tier int64 | complexity_axis stringclasses 0
values | complexity_tier_label stringclasses 0
values | calibration_iteration stringclasses 0
values | is_incumbent_v20 bool 0
classes | is_v21_interpolation bool 0
classes | sol_reward float64 | sol_solved float64 | sol_turns float64 | gemini_evidence stringclasses 0
values | gemini_reward float64 | gemini_solved float64 | gemini_turns float64 | gemini_censor_turn int64 | gemini_censor_type stringclasses 0
values |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
odd ranks | legacy_train | {} | ["static_rank"] | legacy_train | {} | 0 | 0 | 0 | 0 | 16 | 0.538462 | 0.995727 | 3234b7c3c944adadebdae4978abd254376bfddf12d8283b633bd7882365053ca | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_rank_odd | odd ranks | static_rank | return card.rank % 2 == 1 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
hearts only | legacy_train | {} | ["static_suit_color"] | legacy_train | {} | 0 | 0 | 0 | 0 | 13 | 0.25 | 0.811278 | b0ccba9c4aca8995864f4b1e9e293b850ff1afa000536d1c565a105184997025 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_suit_hearts | hearts only | static_suit_color | return card.suit == "hearts" | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
black ranks ace through 7 | legacy_train | {} | ["static_combo"] | legacy_train | {} | 0 | 0 | 0 | 0 | 22 | 0.269231 | 0.840359 | f79c85e8e50b8a8ee32866c9b534e4541710e4f5bb0a66bfcaec112082c46a43 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_combo_black_low | black ranks ace through 7 | static_combo | return card.color == "black" and card.rank <= 7 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
same color as previous card | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 28 | 0.5 | 1 | f8e987bebd21470908a0302e820b4cf7588041e7d5d143ecd13a02cc8947db32 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_same_color | same color as previous card | previous_card | if not mainline:
return True
return card.color == mainline[-1].color | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
colors appear in pairs and each pair changes color | legacy_train | {} | ["pair_position"] | legacy_train | {} | 0 | 0 | 0 | 0 | 56 | 0.5 | 1 | 42106fa1025f6491f388d1b5a34270f7f88667738fb600e1eb0d7471b93d3d4e | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_pair_colors | colors appear in pairs and each pair changes color | pair_position | if not mainline:
return True
if len(mainline) % 2 == 0:
return card.color != mainline[-1].color
return card.color == mainline[-1].color | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
ranks ace through 6 | legacy_train | {} | ["static_rank"] | legacy_train | {} | 0 | 0 | 0 | 0 | 13 | 0.461538 | 0.995727 | 458f450741a1f05c2c404208afa95b5c1a58c938315d8331d0969edb2c4bc5c2 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_rank_low_1_6 | ranks ace through 6 | static_rank | return card.rank <= 6 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
anything except diamonds | legacy_train | {} | ["static_suit_color"] | legacy_train | {} | 0 | 0 | 0 | 0 | 13 | 0.75 | 0.811278 | e23f167d53ade227893076718eddfce18ffbddf28945d754f7718d31542a6c43 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_suit_no_diamonds | anything except diamonds | static_suit_color | return card.suit != "diamonds" | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
red face cards | legacy_train | {} | ["static_combo"] | legacy_train | {} | 0 | 0 | 0 | 0 | 22 | 0.115385 | 0.515947 | af2835388ce614ac03129c8f47e87481d33ef8bed2c8b74af73f7addaad32778 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_combo_red_face | red face cards | static_combo | return card.color == "red" and card.rank >= 11 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
rank parity appears in pairs and each pair changes parity | legacy_train | {} | ["pair_position"] | legacy_train | {} | 0 | 0 | 0 | 0 | 68 | 0.503539 | 0.999964 | d118ee59cf223516fdacadfcab1445b7e54209ebe43726b156acead9238b63b5 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_pair_parity | rank parity appears in pairs and each pair changes parity | pair_position | if not mainline:
return True
if len(mainline) % 2 == 0:
return (card.rank % 2) != (mainline[-1].rank % 2)
return (card.rank % 2) == (mainline[-1].rank % 2) | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
ranks 8 through king | legacy_train | {} | ["static_rank"] | legacy_train | {} | 0 | 0 | 0 | 0 | 13 | 0.461538 | 0.995727 | cb6aaa972f368f7453e4fb7e2a891564e91c073269df66cb025ae2421ad7449f | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_rank_high_8_13 | ranks 8 through king | static_rank | return card.rank >= 8 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
black cards | legacy_train | {} | ["static_suit_color"] | legacy_train | {} | 0 | 0 | 0 | 0 | 13 | 0.5 | 1 | 326143723d615b2a2fda493155578d3974a83a388fed6a0034f974f9449f8318 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_suit_black | black cards | static_suit_color | return card.color == "black" | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
red odd cards and black even cards | legacy_train | {} | ["static_combo"] | legacy_train | {} | 0 | 0 | 0 | 0 | 35 | 0.5 | 1 | 5006276397bba7dff6cc7ea6dd2f8593fdf6f97f1dd2767b95cc7c774fa568be | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_combo_red_odd_black_even | red odd cards and black even cards | static_combo | if card.color == "red":
return card.rank % 2 == 1
return card.rank % 2 == 0 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
same suit as previous card | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 28 | 0.25 | 0.811278 | adb95f4b599d2e09af9adc0842605015f09487d1549416084df9e1edb3a509c0 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_same_suit | same suit as previous card | previous_card | if not mainline:
return True
return card.suit == mainline[-1].suit | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
ranks 4 through 10 | legacy_train | {} | ["static_rank"] | legacy_train | {} | 0 | 0 | 0 | 0 | 15 | 0.538462 | 0.995727 | 196d1583cee34fc43efbb724eea407122372385283ced9ffc2cd5c999e4b4a77 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_rank_middle_4_10 | ranks 4 through 10 | static_rank | return 4 <= card.rank <= 10 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
hearts or clubs | legacy_train | {} | ["static_suit_color"] | legacy_train | {} | 0 | 0 | 0 | 0 | 15 | 0.5 | 1 | 5a92f6544ea4b8f03f0106eecd2345bad734f57a64d87bc1868458b8ecf47f1d | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_suit_hearts_or_clubs | hearts or clubs | static_suit_color | return card.suit in {"hearts", "clubs"} | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
non-prime ranks | legacy_train | {} | ["static_rank"] | legacy_train | {} | 0 | 0 | 0 | 0 | 24 | 0.538462 | 0.995727 | fc056be41c5916c580f34aaee346e5f3870525d99b3ea84e08edba32cd4f2d76 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_rank_non_prime | non-prime ranks | static_rank | primes = {2, 3, 5, 7, 11, 13}
return card.rank not in primes | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
same rank as previous card | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 28 | 0.076923 | 0.391244 | f41ba9d5fdc81324edb472c8f9d78146204b4c9039e264d5dbca93df8a6de908 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_same_rank | same rank as previous card | previous_card | if not mainline:
return True
return card.rank == mainline[-1].rank | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
only kings | legacy_train | {} | ["static_rank"] | legacy_train | {} | 0 | 0 | 0 | 0 | 13 | 0.076923 | 0.391244 | 2cffa89a598b96cf0e1d876dd95b0616da5c58f38a95c5db3545a472f18592a0 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_rank_kings | only kings | static_rank | return card.rank == 13 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
same rank parity as previous card | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 34 | 0.507315 | 0.999846 | 97ab64e066032be6eb1950282e7788f462ca5af9bfe58aef33a35481c0cdc778 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_same_parity | same rank parity as previous card | previous_card | if not mainline:
return True
return (card.rank % 2) == (mainline[-1].rank % 2) | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
rank strictly increases | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 28 | 0.500236 | 1 | c3350a6bdfe24ea35df778cd08dca615cace6f9bae9f26d04f12f3b16b84675c | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_strict_increase | rank strictly increases | previous_card | if not mainline:
return True
return card.rank > mainline[-1].rank | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
rank strictly decreases | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 28 | 0.422841 | 0.982753 | b3b6967478a6fdbf155725cef09b200be5a6a48ab3b357138dbb060c22a2b3a1 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_strict_decrease | rank strictly decreases | previous_card | if not mainline:
return True
return card.rank < mainline[-1].rank | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
rank does not increase | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 28 | 0.499764 | 1 | cb35c74d06419f83a71b9e0e110adf54fe6f44ea22a55d50e45f03a1703cd827 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_non_increasing | rank does not increase | previous_card | if not mainline:
return True
return card.rank <= mainline[-1].rank | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
rank changes by exactly 1 | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 34 | 0.142992 | 0.592021 | f47126404939f0c69075c16581f5fd4a88c3f6adfa755935caa5eb38f096cf79 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_rank_diff_exactly_one | rank changes by exactly 1 | previous_card | if not mainline:
return True
return abs(card.rank - mainline[-1].rank) == 1 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
rank changes by at most 1 | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 34 | 0.219915 | 0.760012 | f7d65655a516ec2a239fdfe2dd8a90d294998e490aa8cb0c0014420d99ce233b | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_rank_diff_at_most_one | rank changes by at most 1 | previous_card | if not mainline:
return True
return abs(card.rank - mainline[-1].rank) <= 1 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
rank changes by at most 2 | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 34 | 0.349693 | 0.933794 | 56301a5776e67b3ef3769df0b24dfa62ca56b462ddde109ca251f81758478e90 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_rank_diff_at_most_two | rank changes by at most 2 | previous_card | if not mainline:
return True
return abs(card.rank - mainline[-1].rank) <= 2 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
rank changes by at least 3 | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 34 | 0.650307 | 0.933794 | 0f41f3d2a26bcb494ad201ba2687e8385eb71264a708965bb4a716182dae40ce | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_rank_diff_at_least_three | rank changes by at least 3 | previous_card | if not mainline:
return True
return abs(card.rank - mainline[-1].rank) >= 3 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
same low/high rank group as previous card | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 34 | 0.507787 | 0.999825 | b32a459009f1ff2ac5a9b3755bdc063d70a4e1c9de8e4a46221bbc8c65f837c3 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_same_low_high_group | same low/high rank group as previous card | previous_card | if not mainline:
return True
return (card.rank >= 8) == (mainline[-1].rank >= 8) | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
different low/high rank group than previous card | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 34 | 0.492213 | 0.999825 | 918ba686b2ad6e1d547f94aca9d0a891761637a76cc19ffb8944963c045f106b | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_different_low_high_group | different low/high rank group than previous card | previous_card | if not mainline:
return True
return (card.rank >= 8) != (mainline[-1].rank >= 8) | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
same face/number status as previous card | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 34 | 0.663521 | 0.92141 | 1b7a98a542c851a6cfdb456ec16505e0f89a0c180687db83e1f8ae1efe2716bc | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_same_face_status | same face/number status as previous card | previous_card | if not mainline:
return True
return (card.rank >= 11) == (mainline[-1].rank >= 11) | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
red previous goes down, black previous goes up | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 60 | 0.533742 | 0.996712 | 8dd0697ac5c808d1fd22793ddb5bdba3040986ee440e869d0e96f87db68207e9 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_red_down_black_up | red previous goes down, black previous goes up | previous_card | if not mainline:
return 5 <= card.rank <= 9
prev = mainline[-1]
if prev.color == "red":
return card.rank <= prev.rank
return card.rank >= prev.rank | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
suits cycle diamonds, clubs, hearts, spades | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 57 | 0.25 | 0.811278 | 758a79c62bf07b611cc260c0089a3eb2babbf197530bb39486ad83a418ba0272 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_suit_cycle_dchs | suits cycle diamonds, clubs, hearts, spades | previous_card | if not mainline:
return True
order = ["diamonds", "clubs", "hearts", "spades"]
idx = order.index(mainline[-1].suit)
return card.suit == order[(idx + 1) % 4] | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
faces require different color, numbers require same color | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 52 | 0.5 | 1 | 7b91e59e8863231f59b4b16522c9d973fc8835707e6c9f73a1785aceadb9d36e | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_face_diff_color_number_same_color | faces require different color, numbers require same color | previous_card | if not mainline:
return True
prev = mainline[-1]
if prev.rank >= 11:
return card.color != prev.color
return card.color == prev.color | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
same suit or same rank parity as previous card | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 51 | 0.630486 | 0.950298 | ca5295f424746c45d4a033948a0a2b43b0a0746946512f21b2388b121d283294 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_share_suit_or_parity | same suit or same rank parity as previous card | previous_card | if not mainline:
return True
prev = mainline[-1]
return card.suit == prev.suit or (card.rank % 2) == (prev.rank % 2) | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
clubs only | legacy_train | {} | ["static_suit_color"] | legacy_train | {} | 0 | 0 | 0 | 0 | 13 | 0.25 | 0.811278 | acc4da855e7675baf3f6f6de4f6c581481b5d4c5619b5a84b2596db3dba4360a | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_suit_clubs | clubs only | static_suit_color | return card.suit == "clubs" | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
diamonds only | legacy_train | {} | ["static_suit_color"] | legacy_train | {} | 0 | 0 | 0 | 0 | 13 | 0.25 | 0.811278 | cb17ca3e89a5a6146798c575e5f7d113cb78b933a184de4dc04a41c92bf44df7 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_suit_diamonds | diamonds only | static_suit_color | return card.suit == "diamonds" | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
no hearts | legacy_train | {} | ["static_suit_color"] | legacy_train | {} | 0 | 0 | 0 | 0 | 13 | 0.75 | 0.811278 | 25b5f4756dd1ec6cc21bdc148aa683e7aeb88b007f5991cb419a280f11884850 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_suit_no_hearts | no hearts | static_suit_color | return card.suit != "hearts" | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
hearts or spades | legacy_train | {} | ["static_suit_color"] | legacy_train | {} | 0 | 0 | 0 | 0 | 15 | 0.5 | 1 | 22ee5f228be0f5f0ceb30210a51670886666d172b5ed980831197291ab8020cb | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_suit_hearts_or_spades | hearts or spades | static_suit_color | return card.suit in {"hearts", "spades"} | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
even ranks up to eight | legacy_train | {} | ["static_rank"] | legacy_train | {} | 0 | 0 | 0 | 0 | 17 | 0.307692 | 0.890492 | 3a94e6cd5cc8afe1328dd9659f494040df2c926ab03f4252932ead37ed56081e | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_rank_even_low_set | even ranks up to eight | static_rank | return card.rank in {2, 4, 6, 8} | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
ranks one to three | legacy_train | {} | ["static_rank"] | legacy_train | {} | 0 | 0 | 0 | 0 | 13 | 0.230769 | 0.77935 | 74961539dbe1b403fe3a0e323720224bb6c2e2536945ed5607fb37e93dc28d74 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_rank_low_1_3 | ranks one to three | static_rank | return card.rank <= 3 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
ranks ten and up | legacy_train | {} | ["static_rank"] | legacy_train | {} | 0 | 0 | 0 | 0 | 13 | 0.307692 | 0.890492 | ffd46360857754094fea80595e58d1739282673442433d612b9d530c5ccc80dc | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_rank_high_10_13 | ranks ten and up | static_rank | return card.rank >= 10 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
ranks seven to eleven | legacy_train | {} | ["static_rank"] | legacy_train | {} | 0 | 0 | 0 | 0 | 15 | 0.384615 | 0.961237 | 4be324651fbceaed8ecd37290755105c203fb0649aa97ab91989879b8d7b5d01 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_rank_band_7_11 | ranks seven to eleven | static_rank | return 7 <= card.rank <= 11 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
black high cards | legacy_train | {} | ["static_combo"] | legacy_train | {} | 0 | 0 | 0 | 0 | 22 | 0.153846 | 0.619382 | 08ad5662070bcf6231e2b23a23b4dae26e3897c937429d76b917798902940ab1 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_combo_black_high | black high cards | static_combo | return card.color == "black" and card.rank >= 10 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
alternate low and mid-high vs previous (threshold five) | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 34 | 0.444549 | 0.99111 | afa7e08c1ab6776eb94e5f190d8d98a8040965e00380aa48bae13db8b62f4468 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_different_mid_group | alternate low and mid-high vs previous (threshold five) | previous_card | if not mainline:
return True
return (card.rank >= 5) != (mainline[-1].rank >= 5) | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
pairs share suit (opposite phase) | legacy_train | {} | ["pair_position"] | legacy_train | {} | 0 | 0 | 0 | 0 | 56 | 0.492331 | 0.99983 | 31714cc5dde0f44442db2fa5fdd19a1c1159e31020180e1f68dd3330da4210f5 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_pair_suits | pairs share suit (opposite phase) | pair_position | if not mainline:
return True
if len(mainline) % 2 == 0:
return card.suit == mainline[-1].suit
return card.suit != mainline[-1].suit | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
pairs share rank (opposite phase) | legacy_train | {} | ["pair_position"] | legacy_train | {} | 0 | 0 | 0 | 0 | 56 | 0.487022 | 0.999514 | 432087e82f0b7f6b1a0d1c66b63183342d7967f6a44aed0670a85da564c6e334 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_pair_ranks | pairs share rank (opposite phase) | pair_position | if not mainline:
return True
if len(mainline) % 2 == 0:
return card.rank == mainline[-1].rank
return card.rank != mainline[-1].rank | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
red even ranks | legacy_train | {} | ["static_combo"] | legacy_train | {} | 0 | 0 | 0 | 0 | 25 | 0.230769 | 0.77935 | 6b4988f91b3ba99be9230763418b8ffbb091ad3a4d6ef6065e17837696e2d9f2 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_rank_even_red | red even ranks | static_combo | return card.color == "red" and card.rank % 2 == 0 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
black even ranks | legacy_train | {} | ["static_combo"] | legacy_train | {} | 0 | 0 | 0 | 0 | 25 | 0.230769 | 0.77935 | 125fb8a35692301fbbb7c57bbfe5d1c5f3337cfa281e7fccd1ae3f019fe8c9fa | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_rank_even_black | black even ranks | static_combo | return card.color == "black" and card.rank % 2 == 0 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
even ranks or kings | legacy_train | {} | ["static_rank"] | legacy_train | {} | 0 | 0 | 0 | 0 | 25 | 0.538462 | 0.995727 | f0f912b8912968ca4ec3f8bc147a74d3bdc11b45cb69ddf223c8ecfea306dcc6 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_rank_even_or_king | even ranks or kings | static_rank | return card.rank % 2 == 0 or card.rank == 13 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
odd primes | legacy_train | {} | ["static_rank"] | legacy_train | {} | 0 | 0 | 0 | 0 | 18 | 0.384615 | 0.961237 | 7d792e7d678827a72903a625d58a5849ddf14a455c649c167a4c1f6a679d06e3 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_rank_odd_primes | odd primes | static_rank | return card.rank in {3, 5, 7, 11, 13} | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
alternate low and high vs previous (threshold ten) | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 34 | 0.394998 | 0.967949 | a3f4f8734dcf6d696b602789ee60bf497601da0f92ab25285fbb89bf4cd6ac06 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_different_high_group_10 | alternate low and high vs previous (threshold ten) | previous_card | if not mainline:
return True
return (card.rank >= 10) != (mainline[-1].rank >= 10) | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
alternate between hearts+clubs and diamonds+spades | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 42 | 0.5 | 1 | 0442debe2aa9fd1bb773ba21cc5923309ff2d0da935237e5a6b2f6e7ca18b5d3 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_group_hc_ds | alternate between hearts+clubs and diamonds+spades | previous_card | if not mainline:
return True
group_a = {"hearts", "clubs"}
return (card.suit in group_a) != (mainline[-1].suit in group_a) | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
after a face card same color, after a number card different color | legacy_train | {} | ["previous_card"] | legacy_train | {} | 0 | 0 | 0 | 0 | 49 | 0.5 | 1 | 647cd8a5c267b323d21968c8075cbc4dc50bab6da15781ccfe2e0de86872e1c8 | 0 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | train_seq_face_same_color_number_diff_color | after a face card same color, after a number card different color | previous_card | if not mainline:
return True
prev = mainline[-1]
if prev.is_face:
return card.color == prev.color
return card.color != prev.color | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's rank parity is 3 modulo 5 | candidate_attribute_count_phase | {"attribute": "parity", "modulus": 5, "residue": 3} | ["global_history"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 8, "history_depth": 16, "period_or_chunk_length": 5} | 16 | 5 | 0 | 2 | 50 | 0.178858 | 0.677572 | d81e460cd6c41064a9467ef8c2dc2ed99a010a2b2bd7caae2a7e861c822a4de1 | 8 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_11a34f4069 | the prior count of the candidate's rank parity is 3 modulo 5 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.rank % 2 == item.rank % 2)
return count % 5 == 3 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's color is 0 modulo 4 | candidate_attribute_count_phase | {"attribute": "color", "modulus": 4, "residue": 0} | ["global_history"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 8, "history_depth": 16, "period_or_chunk_length": 4} | 16 | 4 | 0 | 2 | 44 | 0.254601 | 0.81849 | af751702beab8091455c5ae845bc3dbb2b02c0f18c68554ab15f4147242029ab | 8 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_1305785d8a | the prior count of the candidate's color is 0 modulo 4 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.color == item.color)
return count % 4 == 0 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's face status is 0 modulo 2 | candidate_attribute_count_phase | {"attribute": "face", "modulus": 2, "residue": 0} | ["global_history"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 7, "history_depth": 16, "period_or_chunk_length": 2} | 16 | 2 | 0 | 2 | 44 | 0.578575 | 0.982112 | a262869d1e591681cb0a0916d5cdaf64d3c2629bbef03cbe8381f5b6c1ed8e2d | 7 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_13879dc61b | the prior count of the candidate's face status is 0 modulo 2 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.is_face == item.is_face)
return count % 2 == 0 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's rank parity is 2 modulo 4 | candidate_attribute_count_phase | {"attribute": "parity", "modulus": 4, "residue": 2} | ["global_history"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 8, "history_depth": 16, "period_or_chunk_length": 4} | 16 | 4 | 0 | 2 | 50 | 0.249174 | 0.809967 | 1aa32eef3dc2a0529b9040f38f5c8b33e00b47d56ae803035101e92d0b8af2d7 | 8 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_19a28414b4 | the prior count of the candidate's rank parity is 2 modulo 4 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.rank % 2 == item.rank % 2)
return count % 4 == 2 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's color is 0 modulo 2 | candidate_attribute_count_phase | {"attribute": "color", "modulus": 2, "residue": 0} | ["global_history"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 7, "history_depth": 16, "period_or_chunk_length": 2} | 16 | 2 | 0 | 2 | 44 | 0.52454 | 0.998262 | d3f3563786418ae5e2214a3afbc250dd8de68f398c1fb38fe2cae756e4c92e75 | 7 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_2d018d218c | the prior count of the candidate's color is 0 modulo 2 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.color == item.color)
return count % 2 == 0 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's rank is 0 modulo 3 | candidate_attribute_count_phase | {"attribute": "rank", "modulus": 3, "residue": 0} | ["global_history"] | train_parameterization | {"attribute_cardinality": 13, "branch_count": 0, "designed_level": 8, "history_depth": 16, "period_or_chunk_length": 3} | 16 | 3 | 0 | 13 | 44 | 0.554035 | 0.991559 | 5a7fbe91ef96a998a4d3c491ce8b190af7930bed4bcd3024935c814038a07c72 | 8 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_2dc38c5afc | the prior count of the candidate's rank is 0 modulo 3 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.rank == item.rank)
return count % 3 == 0 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's suit is 2 modulo 3 | candidate_attribute_count_phase | {"attribute": "suit", "modulus": 3, "residue": 2} | ["global_history"] | train_parameterization | {"attribute_cardinality": 4, "branch_count": 0, "designed_level": 7, "history_depth": 16, "period_or_chunk_length": 3} | 16 | 3 | 0 | 4 | 44 | 0.266871 | 0.836939 | da42bb640b144762708df59f8b38868267fd057c9eb0e4dae744c08a713dc5ee | 7 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_32857bfa82 | the prior count of the candidate's suit is 2 modulo 3 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.suit == item.suit)
return count % 3 == 2 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's suit is 0 modulo 3 | candidate_attribute_count_phase | {"attribute": "suit", "modulus": 3, "residue": 0} | ["global_history"] | train_parameterization | {"attribute_cardinality": 4, "branch_count": 0, "designed_level": 7, "history_depth": 16, "period_or_chunk_length": 3} | 16 | 3 | 0 | 4 | 44 | 0.317485 | 0.901625 | 4e5b4358c79c069eedb6ad19408a0dd54162cca9029e8c4354094fdf3c397258 | 7 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_39d7a98752 | the prior count of the candidate's suit is 0 modulo 3 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.suit == item.suit)
return count % 3 == 0 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's color is 3 modulo 4 | candidate_attribute_count_phase | {"attribute": "color", "modulus": 4, "residue": 3} | ["global_history"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 8, "history_depth": 16, "period_or_chunk_length": 4} | 16 | 4 | 0 | 2 | 44 | 0.226994 | 0.772734 | 4b5a7596df58ab6b0f5fea3682e32f13d0a2d8883106a8f4f00055bc97f579ff | 8 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_3c774c825c | the prior count of the candidate's color is 3 modulo 4 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.color == item.color)
return count % 4 == 3 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's face status is 0 modulo 5 | candidate_attribute_count_phase | {"attribute": "face", "modulus": 5, "residue": 0} | ["global_history"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 8, "history_depth": 16, "period_or_chunk_length": 5} | 16 | 5 | 0 | 2 | 44 | 0.304389 | 0.88659 | 72e05b627754d1f1871999b3408cd318c60ec49661ed91cbdbb64131745ee96d | 8 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_3cc8890e94 | the prior count of the candidate's face status is 0 modulo 5 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.is_face == item.is_face)
return count % 5 == 0 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's suit is 0 modulo 2 | candidate_attribute_count_phase | {"attribute": "suit", "modulus": 2, "residue": 0} | ["global_history"] | train_parameterization | {"attribute_cardinality": 4, "branch_count": 0, "designed_level": 7, "history_depth": 16, "period_or_chunk_length": 2} | 16 | 2 | 0 | 4 | 44 | 0.510736 | 0.999667 | cf34149b3c8cde6ef0e9898c347ff5c121e2b42652dedb63895f6947a40a4d31 | 7 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_404a29c0da | the prior count of the candidate's suit is 0 modulo 2 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.suit == item.suit)
return count % 2 == 0 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's rank is 1 modulo 3 | candidate_attribute_count_phase | {"attribute": "rank", "modulus": 3, "residue": 1} | ["global_history"] | train_parameterization | {"attribute_cardinality": 13, "branch_count": 0, "designed_level": 8, "history_depth": 16, "period_or_chunk_length": 3} | 16 | 3 | 0 | 13 | 44 | 0.404436 | 0.973486 | 7deae35255e9216d10bf364777f2d59d04c5f09342c1cc31cee8e82b8e858a3e | 8 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_59fa9488e7 | the prior count of the candidate's rank is 1 modulo 3 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.rank == item.rank)
return count % 3 == 1 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's suit is 1 modulo 2 | candidate_attribute_count_phase | {"attribute": "suit", "modulus": 2, "residue": 1} | ["global_history"] | train_parameterization | {"attribute_cardinality": 4, "branch_count": 0, "designed_level": 7, "history_depth": 16, "period_or_chunk_length": 2} | 16 | 2 | 0 | 4 | 44 | 0.489264 | 0.999667 | 5cadbce7a0110a1aacafa0a1b68c488f8d2f3a065b6d84d799e3b2e4240c8411 | 7 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_6023db9b3c | the prior count of the candidate's suit is 1 modulo 2 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.suit == item.suit)
return count % 2 == 1 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's rank parity is 0 modulo 3 | candidate_attribute_count_phase | {"attribute": "parity", "modulus": 3, "residue": 0} | ["global_history"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 7, "history_depth": 16, "period_or_chunk_length": 3} | 16 | 3 | 0 | 2 | 50 | 0.423313 | 0.982964 | 1134a2fd6305c8a7f48972a622d6eeaeec397e9a121beb2a2b0cb63f2cf04572 | 7 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_6345aaafed | the prior count of the candidate's rank parity is 0 modulo 3 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.rank % 2 == item.rank % 2)
return count % 3 == 0 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's rank parity is 2 modulo 5 | candidate_attribute_count_phase | {"attribute": "parity", "modulus": 5, "residue": 2} | ["global_history"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 8, "history_depth": 16, "period_or_chunk_length": 5} | 16 | 5 | 0 | 2 | 50 | 0.184049 | 0.688855 | 4b65232f4a56d9bba6b701011b4f5d9730f336909ed659cc7bc4b5677d7efcd8 | 8 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_6ebcbbf801 | the prior count of the candidate's rank parity is 2 modulo 5 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.rank % 2 == item.rank % 2)
return count % 5 == 2 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's color is 2 modulo 4 | candidate_attribute_count_phase | {"attribute": "color", "modulus": 4, "residue": 2} | ["global_history"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 8, "history_depth": 16, "period_or_chunk_length": 4} | 16 | 4 | 0 | 2 | 44 | 0.269939 | 0.841377 | 47cf717a722747245f70293f91362560f4029f613606529788474028a3b0759b | 8 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_86953c33f4 | the prior count of the candidate's color is 2 modulo 4 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.color == item.color)
return count % 4 == 2 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's rank is 1 modulo 4 | candidate_attribute_count_phase | {"attribute": "rank", "modulus": 4, "residue": 1} | ["global_history"] | train_parameterization | {"attribute_cardinality": 13, "branch_count": 0, "designed_level": 9, "history_depth": 16, "period_or_chunk_length": 4} | 16 | 4 | 0 | 13 | 44 | 0.378008 | 0.956623 | a6bfd7f1b6fe61481a16b4955d134f864da4340ffc44c4b8157d3265bc44b4c8 | 9 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_951a18971c | the prior count of the candidate's rank is 1 modulo 4 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.rank == item.rank)
return count % 4 == 1 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's rank parity is 3 modulo 4 | candidate_attribute_count_phase | {"attribute": "parity", "modulus": 4, "residue": 3} | ["global_history"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 8, "history_depth": 16, "period_or_chunk_length": 4} | 16 | 4 | 0 | 2 | 50 | 0.181689 | 0.683759 | 8cc162bd5a395b29b4aa1edd5d408e3b96baed4ca82259ac6d2c124917da2de0 | 8 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_b57d1daa7e | the prior count of the candidate's rank parity is 3 modulo 4 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.rank % 2 == item.rank % 2)
return count % 4 == 3 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's face status is 1 modulo 5 | candidate_attribute_count_phase | {"attribute": "face", "modulus": 5, "residue": 1} | ["global_history"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 8, "history_depth": 16, "period_or_chunk_length": 5} | 16 | 5 | 0 | 2 | 44 | 0.174611 | 0.668144 | c1db29f3d4a31edf08b635886f34059b4185f0683bc8acc865c9010b081ab513 | 8 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_bdbf94ac1d | the prior count of the candidate's face status is 1 modulo 5 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.is_face == item.is_face)
return count % 5 == 1 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's rank is 1 modulo 2 | candidate_attribute_count_phase | {"attribute": "rank", "modulus": 2, "residue": 1} | ["global_history"] | train_parameterization | {"attribute_cardinality": 13, "branch_count": 0, "designed_level": 8, "history_depth": 16, "period_or_chunk_length": 2} | 16 | 2 | 0 | 13 | 44 | 0.384615 | 0.961237 | dc89fda154093c84eddf0a4f78b94d49842fe4a40eb53d1ffc5b8ba6f9141875 | 8 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_d5fb8bf89b | the prior count of the candidate's rank is 1 modulo 2 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.rank == item.rank)
return count % 2 == 1 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's suit is 1 modulo 3 | candidate_attribute_count_phase | {"attribute": "suit", "modulus": 3, "residue": 1} | ["global_history"] | train_parameterization | {"attribute_cardinality": 4, "branch_count": 0, "designed_level": 7, "history_depth": 16, "period_or_chunk_length": 3} | 16 | 3 | 0 | 4 | 44 | 0.415644 | 0.979369 | 18e7fd2efafcfcf1c18a543b225b295fae636e5da312f1da065e2df18e91cfd7 | 7 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_d74b2f1193 | the prior count of the candidate's suit is 1 modulo 3 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.suit == item.suit)
return count % 3 == 1 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's rank is 0 modulo 2 | candidate_attribute_count_phase | {"attribute": "rank", "modulus": 2, "residue": 0} | ["global_history"] | train_parameterization | {"attribute_cardinality": 13, "branch_count": 0, "designed_level": 8, "history_depth": 16, "period_or_chunk_length": 2} | 16 | 2 | 0 | 13 | 44 | 0.615385 | 0.961237 | e30f8f0a96ccc0f5c318285629578726e7608ab372c473509d7a1fb4a33a9ccc | 8 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_d8bfad3d7a | the prior count of the candidate's rank is 0 modulo 2 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.rank == item.rank)
return count % 2 == 0 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's face status is 1 modulo 2 | candidate_attribute_count_phase | {"attribute": "face", "modulus": 2, "residue": 1} | ["global_history"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 7, "history_depth": 16, "period_or_chunk_length": 2} | 16 | 2 | 0 | 2 | 44 | 0.421425 | 0.982112 | 8e6d380d0ce0f7cc1206947aaf87263ed0a57aacd51bfa8945c61179987b4948 | 7 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_e15deef5a4 | the prior count of the candidate's face status is 1 modulo 2 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.is_face == item.is_face)
return count % 2 == 1 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's face status is 1 modulo 3 | candidate_attribute_count_phase | {"attribute": "face", "modulus": 3, "residue": 1} | ["global_history"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 7, "history_depth": 16, "period_or_chunk_length": 3} | 16 | 3 | 0 | 2 | 44 | 0.310052 | 0.893233 | f1e2e1f613ba0c3c8f23cab5454aff68eb1fa3235290e2f5684cb5220fe8c3b6 | 7 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_e65ee62fe8 | the prior count of the candidate's face status is 1 modulo 3 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.is_face == item.is_face)
return count % 3 == 1 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's rank parity is 1 modulo 5 | candidate_attribute_count_phase | {"attribute": "parity", "modulus": 5, "residue": 1} | ["global_history"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 8, "history_depth": 16, "period_or_chunk_length": 5} | 16 | 5 | 0 | 2 | 50 | 0.235488 | 0.787457 | 97b96cd6d5f1f65314f904616229caf590f77b91410472ff2cd991b1865e1ab8 | 8 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_e97c6890a5 | the prior count of the candidate's rank parity is 1 modulo 5 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.rank % 2 == item.rank % 2)
return count % 5 == 1 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's face status is 3 modulo 5 | candidate_attribute_count_phase | {"attribute": "face", "modulus": 5, "residue": 3} | ["global_history"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 8, "history_depth": 16, "period_or_chunk_length": 5} | 16 | 5 | 0 | 2 | 44 | 0.186409 | 0.693898 | 3e11b8e0628aa5ce377acf7cb07eccca5e70f13aa5be1d947e6a4c4203f31fd2 | 8 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_ec5003a412 | the prior count of the candidate's face status is 3 modulo 5 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.is_face == item.is_face)
return count % 5 == 3 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's face status is 2 modulo 5 | candidate_attribute_count_phase | {"attribute": "face", "modulus": 5, "residue": 2} | ["global_history"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 8, "history_depth": 16, "period_or_chunk_length": 5} | 16 | 5 | 0 | 2 | 44 | 0.198679 | 0.719277 | ab75a07570b1aeb059d9327ce65a588dfff93af6410f33b310f8c3f5a8d08970 | 8 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_ecc3188d87 | the prior count of the candidate's face status is 2 modulo 5 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.is_face == item.is_face)
return count % 5 == 2 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's color is 4 modulo 5 | candidate_attribute_count_phase | {"attribute": "color", "modulus": 5, "residue": 4} | ["global_history"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 8, "history_depth": 16, "period_or_chunk_length": 5} | 16 | 5 | 0 | 2 | 44 | 0.144172 | 0.595061 | 139081e499fb9ee7ee493cfe4b4de4d3440e95617ebf8160c493cca1043afd50 | 8 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_f1d488b6e4 | the prior count of the candidate's color is 4 modulo 5 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.color == item.color)
return count % 5 == 4 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's color is 2 modulo 3 | candidate_attribute_count_phase | {"attribute": "color", "modulus": 3, "residue": 2} | ["global_history"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 7, "history_depth": 16, "period_or_chunk_length": 3} | 16 | 3 | 0 | 2 | 44 | 0.334356 | 0.919315 | cb6350e9dd83528ccf3a8f447398736ac13e4ca1ae62632f6784d2700387fa55 | 7 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_f709f89fcb | the prior count of the candidate's color is 2 modulo 3 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.color == item.color)
return count % 3 == 2 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's face status is 2 modulo 4 | candidate_attribute_count_phase | {"attribute": "face", "modulus": 4, "residue": 2} | ["global_history"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 8, "history_depth": 16, "period_or_chunk_length": 4} | 16 | 4 | 0 | 2 | 44 | 0.269939 | 0.841377 | f6d470666578a9504b82e62b46db67d979be169e771ada151148956ff5bce242 | 8 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_f8df8ca889 | the prior count of the candidate's face status is 2 modulo 4 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.is_face == item.is_face)
return count % 4 == 2 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
the prior count of the candidate's rank parity is 0 modulo 5 | candidate_attribute_count_phase | {"attribute": "parity", "modulus": 5, "residue": 0} | ["global_history"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 8, "history_depth": 16, "period_or_chunk_length": 5} | 16 | 5 | 0 | 2 | 50 | 0.261444 | 0.828918 | 4a3060f69844dc1161b0148a40fe72dad022b069bdcedc49f191067410503bb5 | 8 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_global_history_candidate_attribute_count_phase_fd7216710a | the prior count of the candidate's rank parity is 0 modulo 5 | global_history | if not mainline:
return True
count = sum(1 for item in mainline if card.rank % 2 == item.rank % 2)
return count % 5 == 0 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
red candidates require same parity; black candidates require same suit, relative to the previous card | candidate_color_gated_transition | {"black_relation": "same_suit", "red_relation": "same_parity"} | ["first_order_transition"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 2, "designed_level": 5, "history_depth": 1, "period_or_chunk_length": 0} | 1 | 0 | 2 | 2 | 58 | 0.388627 | 0.963908 | 520273e86858bab428862b3ac79506b3e9e82e9d889a702fb818f602ea7bec02 | 5 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_first_order_transition_candidate_color_gated_transition_2ca3901093 | red candidates require same parity; black candidates require same suit, relative to the previous card | first_order_transition | if not mainline:
return True
prev = mainline[-1]
if card.color == "red":
return card.rank % 2 == prev.rank % 2
return card.suit == prev.suit | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
red candidates require same parity; black candidates require different parity, relative to the previous card | candidate_color_gated_transition | {"black_relation": "different_parity", "red_relation": "same_parity"} | ["first_order_transition"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 2, "designed_level": 5, "history_depth": 1, "period_or_chunk_length": 0} | 1 | 0 | 2 | 2 | 64 | 0.5 | 1 | 590db012753e2e68c1397d3fb0646d61480da94e075aba6e0a2fdf3f8b9b957f | 5 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_first_order_transition_candidate_color_gated_transition_39d30ae2a6 | red candidates require same parity; black candidates require different parity, relative to the previous card | first_order_transition | if not mainline:
return True
prev = mainline[-1]
if card.color == "red":
return card.rank % 2 == prev.rank % 2
return card.rank % 2 != prev.rank % 2 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
red candidates require different parity; black candidates require same color, relative to the previous card | candidate_color_gated_transition | {"black_relation": "same_color", "red_relation": "different_parity"} | ["first_order_transition"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 2, "designed_level": 5, "history_depth": 1, "period_or_chunk_length": 0} | 1 | 0 | 2 | 2 | 58 | 0.516281 | 0.999235 | ad7d7735bbe34c0ea7a1f601ccc5e8b78dfc2afb1a8170e80323f28e4d7b9643 | 5 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_first_order_transition_candidate_color_gated_transition_3e28d24942 | red candidates require different parity; black candidates require same color, relative to the previous card | first_order_transition | if not mainline:
return True
prev = mainline[-1]
if card.color == "red":
return card.rank % 2 != prev.rank % 2
return card.color == prev.color | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
red candidates require different suit; black candidates require same parity, relative to the previous card | candidate_color_gated_transition | {"black_relation": "same_parity", "red_relation": "different_suit"} | ["first_order_transition"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 2, "designed_level": 5, "history_depth": 1, "period_or_chunk_length": 0} | 1 | 0 | 2 | 2 | 58 | 0.638627 | 0.943817 | 958cafcd19fe6d14a8d300e3f76e18615b5ffdd287bf2a84119d2010fec595fd | 5 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_first_order_transition_candidate_color_gated_transition_504406a04a | red candidates require different suit; black candidates require same parity, relative to the previous card | first_order_transition | if not mainline:
return True
prev = mainline[-1]
if card.color == "red":
return card.suit != prev.suit
return card.rank % 2 == prev.rank % 2 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
red candidates require same color; black candidates require different suit, relative to the previous card | candidate_color_gated_transition | {"black_relation": "different_suit", "red_relation": "same_color"} | ["first_order_transition"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 2, "designed_level": 5, "history_depth": 1, "period_or_chunk_length": 0} | 1 | 0 | 2 | 2 | 52 | 0.595092 | 0.973749 | e79fb349e56e8bbe057839de518aa60a6a1ef98b8cc60b9575273b8fcd57a6b1 | 5 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_first_order_transition_candidate_color_gated_transition_634d02e033 | red candidates require same color; black candidates require different suit, relative to the previous card | first_order_transition | if not mainline:
return True
prev = mainline[-1]
if card.color == "red":
return card.color == prev.color
return card.suit != prev.suit | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
red candidates require same color; black candidates require lower rank, relative to the previous card | candidate_color_gated_transition | {"black_relation": "rank_down", "red_relation": "same_color"} | ["first_order_transition"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 2, "designed_level": 5, "history_depth": 1, "period_or_chunk_length": 0} | 1 | 0 | 2 | 2 | 52 | 0.441482 | 0.990097 | 6da237a5d94accf65dd7b2d2e14ca3b1af67aba6a1411679cc6504c402aa03ef | 5 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_first_order_transition_candidate_color_gated_transition_9770b24cf6 | red candidates require same color; black candidates require lower rank, relative to the previous card | first_order_transition | if not mainline:
return True
prev = mainline[-1]
if card.color == "red":
return card.color == prev.color
return card.rank < prev.rank | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
red candidates require same suit; black candidates require different parity, relative to the previous card | candidate_color_gated_transition | {"black_relation": "different_parity", "red_relation": "same_suit"} | ["first_order_transition"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 2, "designed_level": 5, "history_depth": 1, "period_or_chunk_length": 0} | 1 | 0 | 2 | 2 | 58 | 0.361373 | 0.943817 | 9b83262e5f0801522fa7210b0e14451d352bd7fb3b270862b1725fc4dbc1f4b2 | 5 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_first_order_transition_candidate_color_gated_transition_9edcc4ea33 | red candidates require same suit; black candidates require different parity, relative to the previous card | first_order_transition | if not mainline:
return True
prev = mainline[-1]
if card.color == "red":
return card.suit == prev.suit
return card.rank % 2 != prev.rank % 2 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
red candidates require different parity; black candidates require lower rank, relative to the previous card | candidate_color_gated_transition | {"black_relation": "rank_down", "red_relation": "different_parity"} | ["first_order_transition"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 2, "designed_level": 5, "history_depth": 1, "period_or_chunk_length": 0} | 1 | 0 | 2 | 2 | 58 | 0.457763 | 0.994846 | f806e26bb184bfe6ce1496775daaa7af61d75c33ac249dcdf5f7ccf9601f4d9b | 5 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_first_order_transition_candidate_color_gated_transition_af0a361160 | red candidates require different parity; black candidates require lower rank, relative to the previous card | first_order_transition | if not mainline:
return True
prev = mainline[-1]
if card.color == "red":
return card.rank % 2 != prev.rank % 2
return card.rank < prev.rank | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
red candidates require higher rank; black candidates require same color, relative to the previous card | candidate_color_gated_transition | {"black_relation": "same_color", "red_relation": "rank_up"} | ["first_order_transition"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 2, "designed_level": 5, "history_depth": 1, "period_or_chunk_length": 0} | 1 | 0 | 2 | 2 | 52 | 0.520057 | 0.998839 | 6053923b345d19c3f12c2350a6581ff18bba88f6f445af900468d3ea207d8512 | 5 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_first_order_transition_candidate_color_gated_transition_bb5810f3e2 | red candidates require higher rank; black candidates require same color, relative to the previous card | first_order_transition | if not mainline:
return True
prev = mainline[-1]
if card.color == "red":
return card.rank > prev.rank
return card.color == prev.color | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
red candidates require different color; black candidates require higher rank, relative to the previous card | candidate_color_gated_transition | {"black_relation": "rank_up", "red_relation": "different_color"} | ["first_order_transition"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 2, "designed_level": 5, "history_depth": 1, "period_or_chunk_length": 0} | 1 | 0 | 2 | 2 | 52 | 0.520057 | 0.998839 | 33f0fc54868a313e2f1526ca2e6ca2806c170e1a04cb4e9dd8382fd5f45529e6 | 5 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_first_order_transition_candidate_color_gated_transition_d8efd17dea | red candidates require different color; black candidates require higher rank, relative to the previous card | first_order_transition | if not mainline:
return True
prev = mainline[-1]
if card.color == "red":
return card.color != prev.color
return card.rank > prev.rank | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
different color and different parity | color_and_parity | {"color_same": false, "parity_same": false} | ["first_order_transition"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 3, "history_depth": 1, "period_or_chunk_length": 0} | 1 | 0 | 0 | 2 | 51 | 0.246343 | 0.80543 | 87f5978bc2d7affa248d81cac8813f745afdbea11d4876962391e06adf3e1ed8 | 3 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_first_order_transition_color_and_parity_a5cf1dd5b0 | different color and different parity | first_order_transition | if not mainline:
return True
prev = mainline[-1]
return card.color != prev.color and card.rank % 2 != prev.rank % 2 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
different color and same parity | color_and_parity | {"color_same": false, "parity_same": true} | ["first_order_transition"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 0, "designed_level": 3, "history_depth": 1, "period_or_chunk_length": 0} | 1 | 0 | 0 | 2 | 51 | 0.253657 | 0.817024 | ddb53181b3c7e6149fd2d31d26517182c9083f0166b9820a56e55342ef317a9c | 3 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_first_order_transition_color_and_parity_b6327a134c | different color and same parity | first_order_transition | if not mainline:
return True
prev = mainline[-1]
return card.color != prev.color and card.rank % 2 == prev.rank % 2 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
red cards with rank at most 6 | color_rank_threshold | {"color": "red", "operator": "<=", "threshold": 6} | ["static_predicate"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 1, "designed_level": 2, "history_depth": 0, "period_or_chunk_length": 0} | 0 | 0 | 1 | 2 | 22 | 0.230769 | 0.77935 | 6a6103be8f9fcc6a9184a6d70484920a580baf4bbe1f0dbf397ead9ed0bbb2f2 | 2 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_static_predicate_color_rank_threshold_256c8160a9 | red cards with rank at most 6 | static_predicate | return card.color == "red" and card.rank <= 6 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
red cards with rank at most 4 | color_rank_threshold | {"color": "red", "operator": "<=", "threshold": 4} | ["static_predicate"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 1, "designed_level": 2, "history_depth": 0, "period_or_chunk_length": 0} | 0 | 0 | 1 | 2 | 22 | 0.153846 | 0.619382 | 99f5eed2647ca864dbe6aeee877dc33a8dc39de6538bec05680c76583926071a | 2 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_static_predicate_color_rank_threshold_4081b9d77c | red cards with rank at most 4 | static_predicate | return card.color == "red" and card.rank <= 4 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
red cards with rank at least 4 | color_rank_threshold | {"color": "red", "operator": ">=", "threshold": 4} | ["static_predicate"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 1, "designed_level": 2, "history_depth": 0, "period_or_chunk_length": 0} | 0 | 0 | 1 | 2 | 22 | 0.384615 | 0.961237 | 5ce5f023ea03e33bc42095c96c4106dccf3f80225ca9585602241c063583ad0e | 2 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_static_predicate_color_rank_threshold_50a9928557 | red cards with rank at least 4 | static_predicate | return card.color == "red" and card.rank >= 4 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
red cards with rank at least 10 | color_rank_threshold | {"color": "red", "operator": ">=", "threshold": 10} | ["static_predicate"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 1, "designed_level": 2, "history_depth": 0, "period_or_chunk_length": 0} | 0 | 0 | 1 | 2 | 22 | 0.153846 | 0.619382 | 7f17935d79d08840b5b5ccf834558c5ea2e07125d08f7a4813858028cf30180d | 2 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_static_predicate_color_rank_threshold_7bc8d07b66 | red cards with rank at least 10 | static_predicate | return card.color == "red" and card.rank >= 10 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
black cards with rank at most 11 | color_rank_threshold | {"color": "black", "operator": "<=", "threshold": 11} | ["static_predicate"] | train_parameterization | {"attribute_cardinality": 2, "branch_count": 1, "designed_level": 2, "history_depth": 0, "period_or_chunk_length": 0} | 0 | 0 | 1 | 2 | 22 | 0.423077 | 0.982859 | 3143c4636fea7d4c63dc2761316743d137d70eb97ca262358e2593f963e1eef7 | 2 | -1 | -1 | uncalibrated | [] | v2.1-frontier-calibrated-100turn-20260812 | cal_static_predicate_color_rank_threshold_a46bbf3a4f | black cards with rank at most 11 | static_predicate | return card.color == "black" and card.rank <= 11 | eleusis-100-v11 | model_balanced_mean_reward | 100 | 30 | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null |
End of preview. Expand in Data Studio
Eleusis Calibrated Rules — 100-turn reward calibration
A calibrated rule dataset for the single-player Eleusis inductive-reasoning environment. It extends the 26-rule Hugging Face benchmark with controlled static, transition, conditional, periodic, chunk, higher-order history, global history, and compositional rule families.
Source benchmark: Hugging Face Eleusis.
Dataset version: v2.1-frontier-calibrated-100turn-20260812
Protocol: eleusis-100-v11
The structural, GPT Sol difficulty, family, and long-horizon readiness gates passed.
Gemini confirmation is incomplete: 11 exact endpoints, 11 right-censored prefixes, and 10 not-run rules. Censored traces are not scored as failures.
Splits
{ "train": 959, "validation": 289, "test": 32, "legacy_test": 26 }
- train: parameterized training rules.
- validation: held-out parameterizations and compositions.
- test: 32-rule, family-balanced reward-calibration suite (4 per family, one at each declared complexity tier).
- legacy_test: unchanged 26-rule Hugging Face benchmark.
Calibration target and result
- GPT-5.6 Sol solve@100: 0.500
- GPT-5.6 Sol normalized reward: 0.253
- Family mean-reward spread: 0.292
- Exact per-rule calibration panel stored in the test rows: openai/gpt-5.6-sol
The 100-turn test was selected by normalized reward and solve rate, with equal family quotas and explicit within-family complexity tiers. Every family keeps at least one rule GPT Sol did not solve at 100 turns, and every family reward is below 0.5. Gemini construction evidence is included with exact/censored/missing status; no missing or censored episode is imputed into its performance. Detailed per-rule results, rule-bootstrap intervals, costs, and visualizations are included with the calibration artifacts. The final report explains what was built, how the calibration was run, and what the family and complexity results imply.
Family coverage in the test split is:
{
"static_predicate": 4,
"first_order_transition": 4,
"conditional_transition": 4,
"periodic_cycle": 4,
"chunk_run": 4,
"higher_order_history": 4,
"global_history": 4,
"compositional_hybrid": 4
}
Core fields
Every split retains rule_id, label, family, and code. Additional fields describe template lineage, semantic complexity, observability, empirical difficulty, and dataset version.
Reproducibility
Use the published Eleusis taskset with the test split, 100 turns, temperature 0.7, and 4,096 maximum completion tokens per model call. The engine preserves the legacy deal prefix and appends deterministic reserve shoes only when the longer game needs more cards. The legacy split preserves historical comparison with the source benchmark.
The default harness retains the full transcript and does not compact it. A provider-neutral user heartbeat follows each tool result, and one continuation is allowed only after a provider-reported completion-length cutoff. The calibration report contains measured context growth and the failure policy for models with smaller context windows.
Calibration is empirical, selection-conditioned, and protocol-specific. Model updates, sampling variance, or prompt changes can shift absolute scores; use the included per-rule metadata and calibration artifacts to re-estimate difficulty as the frontier changes. Fresh models and fresh deal seeds should be used for the confirmatory full-panel comparison.
Repository: https://huggingface.co/datasets/nph4rd/eleusis-calibrated-rules
- Downloads last month
- 84