karti-ai
2ef62c5670
32 recorded rollouts, 8 seeds x 4 agents, every one replaying through the Python engine to a delta of exactly 0. arm solved economy consistency total base-off 0/8 0.000 0.122 0.0244 base-on 4/8 0.469 0.479 0.4865 cautious 8/8 0.944 1.000 0.9831 solver 8/8 1.000 0.719 0.9437 Thinking-off solves none of eight. The same model on the same seeds, sampled with thinking on, solves four. That is the headline, it is ours, and it needed no training — which is also why the run is labelled an intervention and names what was done to it, so a sampling change can never read as a training result. `cautious` is new and it exists to make the reward editor honest. Until now the page invited you to move a slider and watch the ranking change, and no slider changed anything, because the solver dominated a model that solved nothing. A candidate-only player — most informative guess among words that could still win — takes consistency outright and pays for it in turns. Now: only speed matters solver 0.9859 cautious 0.9606 -> solver shipped weights solver 0.9437 cautious 0.9831 -> cautious punish contradictions solver 0.8312 cautious 0.9972 -> cautious The ranking really does flip, on recorded data, with one slider. The first version of this arm picked the alphabetically-first candidate and opened on 'abaci', which made the policy the counterweight exists to reward look like a straw man. 190 contract checks pass. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019mt6sHQHEnEYrJZvoMCJSB
301 lines
8.3 KiB
JSON
301 lines
8.3 KiB
JSON
{
|
|
"wordle": [
|
|
{
|
|
"id": "base-off-s0",
|
|
"label": "Out of the box",
|
|
"path": "/traces/wordle/base-off-s0.json",
|
|
"kind": "recorded",
|
|
"model": "brain-qwen38-dspark",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 0
|
|
},
|
|
{
|
|
"id": "base-off-s1",
|
|
"label": "Out of the box",
|
|
"path": "/traces/wordle/base-off-s1.json",
|
|
"kind": "recorded",
|
|
"model": "brain-qwen38-dspark",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 1
|
|
},
|
|
{
|
|
"id": "base-off-s2",
|
|
"label": "Out of the box",
|
|
"path": "/traces/wordle/base-off-s2.json",
|
|
"kind": "recorded",
|
|
"model": "brain-qwen38-dspark",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 2
|
|
},
|
|
{
|
|
"id": "base-off-s3",
|
|
"label": "Out of the box",
|
|
"path": "/traces/wordle/base-off-s3.json",
|
|
"kind": "recorded",
|
|
"model": "brain-qwen38-dspark",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 3
|
|
},
|
|
{
|
|
"id": "base-off-s4",
|
|
"label": "Out of the box",
|
|
"path": "/traces/wordle/base-off-s4.json",
|
|
"kind": "recorded",
|
|
"model": "brain-qwen38-dspark",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 4
|
|
},
|
|
{
|
|
"id": "base-off-s5",
|
|
"label": "Out of the box",
|
|
"path": "/traces/wordle/base-off-s5.json",
|
|
"kind": "recorded",
|
|
"model": "brain-qwen38-dspark",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 5
|
|
},
|
|
{
|
|
"id": "base-off-s6",
|
|
"label": "Out of the box",
|
|
"path": "/traces/wordle/base-off-s6.json",
|
|
"kind": "recorded",
|
|
"model": "brain-qwen38-dspark",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 6
|
|
},
|
|
{
|
|
"id": "base-off-s7",
|
|
"label": "Out of the box",
|
|
"path": "/traces/wordle/base-off-s7.json",
|
|
"kind": "recorded",
|
|
"model": "brain-qwen38-dspark",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 7
|
|
},
|
|
{
|
|
"id": "base-on-s0",
|
|
"label": "Allowed to think",
|
|
"path": "/traces/wordle/base-on-s0.json",
|
|
"kind": "intervened",
|
|
"model": "brain-qwen38-dspark",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 0,
|
|
"intervention": "Same model, same seeds, sampled with thinking enabled. No training, no fine-tuning."
|
|
},
|
|
{
|
|
"id": "base-on-s1",
|
|
"label": "Allowed to think",
|
|
"path": "/traces/wordle/base-on-s1.json",
|
|
"kind": "intervened",
|
|
"model": "brain-qwen38-dspark",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 1,
|
|
"intervention": "Same model, same seeds, sampled with thinking enabled. No training, no fine-tuning."
|
|
},
|
|
{
|
|
"id": "base-on-s2",
|
|
"label": "Allowed to think",
|
|
"path": "/traces/wordle/base-on-s2.json",
|
|
"kind": "intervened",
|
|
"model": "brain-qwen38-dspark",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 2,
|
|
"intervention": "Same model, same seeds, sampled with thinking enabled. No training, no fine-tuning."
|
|
},
|
|
{
|
|
"id": "base-on-s3",
|
|
"label": "Allowed to think",
|
|
"path": "/traces/wordle/base-on-s3.json",
|
|
"kind": "intervened",
|
|
"model": "brain-qwen38-dspark",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 3,
|
|
"intervention": "Same model, same seeds, sampled with thinking enabled. No training, no fine-tuning."
|
|
},
|
|
{
|
|
"id": "base-on-s4",
|
|
"label": "Allowed to think",
|
|
"path": "/traces/wordle/base-on-s4.json",
|
|
"kind": "intervened",
|
|
"model": "brain-qwen38-dspark",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 4,
|
|
"intervention": "Same model, same seeds, sampled with thinking enabled. No training, no fine-tuning."
|
|
},
|
|
{
|
|
"id": "base-on-s5",
|
|
"label": "Allowed to think",
|
|
"path": "/traces/wordle/base-on-s5.json",
|
|
"kind": "intervened",
|
|
"model": "brain-qwen38-dspark",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 5,
|
|
"intervention": "Same model, same seeds, sampled with thinking enabled. No training, no fine-tuning."
|
|
},
|
|
{
|
|
"id": "base-on-s6",
|
|
"label": "Allowed to think",
|
|
"path": "/traces/wordle/base-on-s6.json",
|
|
"kind": "intervened",
|
|
"model": "brain-qwen38-dspark",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 6,
|
|
"intervention": "Same model, same seeds, sampled with thinking enabled. No training, no fine-tuning."
|
|
},
|
|
{
|
|
"id": "base-on-s7",
|
|
"label": "Allowed to think",
|
|
"path": "/traces/wordle/base-on-s7.json",
|
|
"kind": "intervened",
|
|
"model": "brain-qwen38-dspark",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 7,
|
|
"intervention": "Same model, same seeds, sampled with thinking enabled. No training, no fine-tuning."
|
|
},
|
|
{
|
|
"id": "solver-s0",
|
|
"label": "Best-known play",
|
|
"path": "/traces/wordle/solver-s0.json",
|
|
"kind": "generated",
|
|
"model": "entropy-solver",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 0
|
|
},
|
|
{
|
|
"id": "solver-s1",
|
|
"label": "Best-known play",
|
|
"path": "/traces/wordle/solver-s1.json",
|
|
"kind": "generated",
|
|
"model": "entropy-solver",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 1
|
|
},
|
|
{
|
|
"id": "solver-s2",
|
|
"label": "Best-known play",
|
|
"path": "/traces/wordle/solver-s2.json",
|
|
"kind": "generated",
|
|
"model": "entropy-solver",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 2
|
|
},
|
|
{
|
|
"id": "solver-s3",
|
|
"label": "Best-known play",
|
|
"path": "/traces/wordle/solver-s3.json",
|
|
"kind": "generated",
|
|
"model": "entropy-solver",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 3
|
|
},
|
|
{
|
|
"id": "solver-s4",
|
|
"label": "Best-known play",
|
|
"path": "/traces/wordle/solver-s4.json",
|
|
"kind": "generated",
|
|
"model": "entropy-solver",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 4
|
|
},
|
|
{
|
|
"id": "solver-s5",
|
|
"label": "Best-known play",
|
|
"path": "/traces/wordle/solver-s5.json",
|
|
"kind": "generated",
|
|
"model": "entropy-solver",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 5
|
|
},
|
|
{
|
|
"id": "solver-s6",
|
|
"label": "Best-known play",
|
|
"path": "/traces/wordle/solver-s6.json",
|
|
"kind": "generated",
|
|
"model": "entropy-solver",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 6
|
|
},
|
|
{
|
|
"id": "solver-s7",
|
|
"label": "Best-known play",
|
|
"path": "/traces/wordle/solver-s7.json",
|
|
"kind": "generated",
|
|
"model": "entropy-solver",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 7
|
|
},
|
|
{
|
|
"id": "cautious-s0",
|
|
"label": "Never wastes a guess",
|
|
"path": "/traces/wordle/cautious-s0.json",
|
|
"kind": "generated",
|
|
"model": "candidate-only-solver",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 0
|
|
},
|
|
{
|
|
"id": "cautious-s1",
|
|
"label": "Never wastes a guess",
|
|
"path": "/traces/wordle/cautious-s1.json",
|
|
"kind": "generated",
|
|
"model": "candidate-only-solver",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 1
|
|
},
|
|
{
|
|
"id": "cautious-s2",
|
|
"label": "Never wastes a guess",
|
|
"path": "/traces/wordle/cautious-s2.json",
|
|
"kind": "generated",
|
|
"model": "candidate-only-solver",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 2
|
|
},
|
|
{
|
|
"id": "cautious-s3",
|
|
"label": "Never wastes a guess",
|
|
"path": "/traces/wordle/cautious-s3.json",
|
|
"kind": "generated",
|
|
"model": "candidate-only-solver",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 3
|
|
},
|
|
{
|
|
"id": "cautious-s4",
|
|
"label": "Never wastes a guess",
|
|
"path": "/traces/wordle/cautious-s4.json",
|
|
"kind": "generated",
|
|
"model": "candidate-only-solver",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 4
|
|
},
|
|
{
|
|
"id": "cautious-s5",
|
|
"label": "Never wastes a guess",
|
|
"path": "/traces/wordle/cautious-s5.json",
|
|
"kind": "generated",
|
|
"model": "candidate-only-solver",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 5
|
|
},
|
|
{
|
|
"id": "cautious-s6",
|
|
"label": "Never wastes a guess",
|
|
"path": "/traces/wordle/cautious-s6.json",
|
|
"kind": "generated",
|
|
"model": "candidate-only-solver",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 6
|
|
},
|
|
{
|
|
"id": "cautious-s7",
|
|
"label": "Never wastes a guess",
|
|
"path": "/traces/wordle/cautious-s7.json",
|
|
"kind": "generated",
|
|
"model": "candidate-only-solver",
|
|
"capturedAt": "2026-08-28",
|
|
"seed": 7
|
|
}
|
|
]
|
|
}
|