Evals
Runs of the physicist over a set of corpus questions. Each run is a real question, stamped with the agent version that solved it.
- pr72 v4 branch rerun 3 done 14 Oct 15:28 · only · 4 runs stated-pi-squared · 5f5be23 100% 4/4 a_stone_on_a_string caesium_s_stopping_potential how_long_to_stop pushed_two_ways
- pr72 v4 main rerun 3 done 14 Oct 15:19 · only · 4 runs bd007a8 75% 3/4 a_stone_on_a_string caesium_s_stopping_potential how_long_to_stop pushed_two_ways
- pr72 v4 branch rerun 2 done 14 Oct 15:16 · only · 4 runs stated-pi-squared · 5f5be23 50% 2/4 a_stone_on_a_string caesium_s_stopping_potential how_long_to_stop pushed_two_ways
- pr72 v4 main rerun 2 done 14 Oct 15:12 · only · 4 runs bd007a8 75% 3/4 a_stone_on_a_string caesium_s_stopping_potential how_long_to_stop pushed_two_ways
- pr72 v4 branch rerun 1 done 14 Oct 15:12 · only · 4 runs stated-pi-squared · 5f5be23 100% 4/4 a_stone_on_a_string caesium_s_stopping_potential how_long_to_stop pushed_two_ways
- pr72 v4 main rerun 1 done 14 Oct 15:09 · only · 4 runs bd007a8 50% 2/4 a_stone_on_a_string caesium_s_stopping_potential how_long_to_stop pushed_two_ways
- backend edge-thanveer done 14 Oct 10:10 · six · 18 runs bdd945f 61% 11/18 a_hot_copper_block_on_ice a_stone_on_a_string caesium_s_stopping_potential how_long_to_stop pushed_two_ways two_parallel_wires
- backend edge done 14 Oct 09:57 · six · 18 runs bdd945f 72% 13/18 a_hot_copper_block_on_ice a_stone_on_a_string caesium_s_stopping_potential how_long_to_stop pushed_two_ways two_parallel_wires
- main bdd945f six done 14 Oct 09:01 · six · 18 runs bdd945f 39% 7/18 a_hot_copper_block_on_ice a_stone_on_a_string caesium_s_stopping_potential how_long_to_stop pushed_two_ways two_parallel_wires
- multi-llm main done 14 Oct 05:32 · six · 14 runs f0f5267 71% 5/7 f0f5267 60% 3/5 a_hot_copper_block_on_ice a_stone_on_a_string caesium_s_stopping_potential how_long_to_stop pushed_two_ways two_parallel_wires
- multi-llm branch done 14 Oct 02:46 · six · 18 runs multi-llm-servers · 2b11b97 72% 13/18 a_hot_copper_block_on_ice a_stone_on_a_string caesium_s_stopping_potential how_long_to_stop pushed_two_ways two_parallel_wires