| model_family,canonical_train_run_id,hub_slug,checkpoint_step,validation_examples,target_prefix_exact_match,retained_sequence_exact_match,selected_checkpoint,selection_namespace,selection_phase,selection_metric,selection_tie_break,source_history_path,source_predictions_sha256,source_result_sha256,postprocess_acceptance_status | |
| qwen,validation_exact_v1/experiments/capacity_study/train/fixed_linear/seed1234_train_d91dd9183c9c,qwen-capacity-fixed-linear-t8-s1234,17000,400,0.2,0.2,false,cache_mask_modes_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_modes_v1/production/corrected_validation_histories.csv,565f6f4bc49512a4e35eaca5e6f2d3a0b1f13bd8ba7e151ca8ead9a9e04ca3de,54f451c6fd19217388a713ac4556e362ce39315c4d5677d0924154f4fd42667f,PASS | |
| qwen,validation_exact_v1/experiments/capacity_study/train/fixed_linear/seed1234_train_d91dd9183c9c,qwen-capacity-fixed-linear-t8-s1234,25000,400,0.2025,0.2025,true,cache_mask_modes_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_modes_v1/production/corrected_validation_histories.csv,cc42a1b3815f66465f5d8368282a24075279e141406c2fb9b090557d11b6ae81,f341dcfab9f74a2b2bd0351c68b5bf3bb5276d27222b173ef9eb91caa576d9e6,PASS | |
| qwen,validation_exact_v1/experiments/capacity_study/train/fixed_residual_mlp/seed1234_train_922bc16f958b,qwen-capacity-fixed-residual-mlp-t8-s1234,18000,400,0.16,0.16,false,cache_mask_modes_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_modes_v1/production/corrected_validation_histories.csv,3bc51a7dc89f6fb3a90912704d6cffa204877b0418553c8699828686e80b23a7,ef7096a250d88ea7cfa5d2e5880b78e9885b9ec9d0cb767bbc99ef7ef4264638,PASS | |
| qwen,validation_exact_v1/experiments/capacity_study/train/fixed_residual_mlp/seed1234_train_922bc16f958b,qwen-capacity-fixed-residual-mlp-t8-s1234,22000,400,0.165,0.165,true,cache_mask_modes_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_modes_v1/production/corrected_validation_histories.csv,5163912a1c7c1fc99c26fed886394dd1504f03c4f6d2883ec91d1050b7e5bb66,46fafd6caec120433aea7ca8b83f7751f8f5e0cf05c052788e0305e127a607db,PASS | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,1000,400,0,0.0,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,305322fb45ca1e6767605e695f0c1882285057d2513db194fa52cb777dd3bd87,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,2000,400,0,0.0,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,d5e3cc6428a38acbaa1264afc8b1b84d74e7eb766c33637e4abec10d5bb1b209,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,3000,400,0.06,0.0025,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,1946bf95d2743490a4512d2fd996f7048ca8505e7a2af057e1a38716a6bb2a21,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,4000,400,0.1625,0.0175,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,2821eb93537d7a1b9c200923b630430c75a13ae623a1c28b179c6f87650ad8fb,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,5000,400,0.21,0.025,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,ca11a148f15b839de53c9c7884921c4cf135130f800d4f9c2a4643b0d12f96b9,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,6000,400,0.2825,0.0475,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,6d3cc8b3cd0e09448386a5198f47b872a92b99cbd82896278e939c007be614b5,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,7000,400,0.3075,0.05,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,24a13f07a4bb7b293ce1d66bf2ec490d7aecc1e3bca4804320bc00423ea1b8df,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,8000,400,0.33,0.0625,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,6a61619efe56f72fbf7e40fa1c73c36765f4096d6082689a513604d2c008cbef,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,9000,400,0.345,0.065,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,e6711c7b4fd4ce418b8a3f6f78bcd85ada535f79352c3aae1fd46a37a6381ef1,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,10000,400,0.355,0.07,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,01d5dc75409f8b9358e8f2f37f3463d55aae5f5a8231be76ea448494efc3c4e9,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,11000,400,0.3525,0.065,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,ed3e8a0c5131136aaa82bbb8aee59ecdc7cfc5e37491d24d40d1ba8e3df63058,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,12000,400,0.3625,0.07,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,876f243ce4a8d9cf7c41ffc26928a689ca6da3fe34789b0afc61cf0710a03aa7,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,13000,400,0.38,0.0675,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,fa5c9dcc75ded72c421b0172689bf302f8d5fb963885fe8d9bc3c38488d2c0af,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,14000,400,0.375,0.065,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,ec3fa1e8a5753c98249d93635f31cdd23362f641cd32b81b2d2a66ed25b5ac30,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,15000,400,0.3875,0.0675,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,b102cf8f9a63ce75b66cdfb61c9a9561742952e890b57e140e3170fbd0a45850,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,16000,400,0.3925,0.0825,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,04bb0fc508c9af62224163bff9c2181b861a4c2d4ef9216b9d6fa8d22ba6270e,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,17000,400,0.4125,0.09,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,43bf8c3a731f71ae729ce6e5cae0192550a6860e888daa4315ca003a7a08806c,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,18000,400,0.3825,0.08,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,2d02225eb9d7478966378dec09f0ecf445033915a09f08936921825c0d3d5101,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,19000,400,0.39,0.07,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,cc50e6b99d8da71c2baaf0d74d6ae16c283471bc2762807b25e718e1165dc667,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,20000,400,0.3975,0.075,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,f615ae5785914ab37816e43cbbe26423524a1b911b4ff5dd4b43f4252a47b053,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,21000,400,0.415,0.0875,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,bfe685f6339a0f5de4de48d3ab0494ba21a4b60fa93e647bbb3aa8ff71e3b188,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,22000,400,0.3975,0.085,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,72c6121c833a573a5a537974fcb6402097dd1bb3f519ff6e4529f56f11da4aa8,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,23000,400,0.39,0.0775,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,8cd781d95a7429d20c89abc15936d72dc31b67174ae7a0b75fe5014f25475140,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,24000,400,0.4,0.0775,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,ef621c3dbb1cc6b831586769f31b607644d3553ab7b5faa470d8c25a3182dbc9,, | |
| qwen,by_representation/full_hidden_sequence/length_scaling/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,qwen-length-full-t32-s1234,25000,400,0.42,0.0825,true,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,e503d971301544f61b2064aa5306d2556419312de1ef2ed2c653bc5f3e1682a7,, | |
| qwen,by_representation/full_hidden_sequence/prefix_components/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_prefix_components_hidden_sequence_t8_final_prefix_direct_only_seed1234,qwen-prefix-direct-only-t8-s1234,22000,400,0.3025,0.3025,false,cache_mask_modes_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_modes_v1/production/corrected_validation_histories.csv,4fe8d84036e7513ef3be3d5ccafea13682a3e1c057ab9fde9d7718273104eb3a,ca51f4389698c88fbe8051dbff3fd3716e6ac752f0a252956d9ea474bd9fb70a,PASS | |
| qwen,by_representation/full_hidden_sequence/prefix_components/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_prefix_components_hidden_sequence_t8_final_prefix_direct_only_seed1234,qwen-prefix-direct-only-t8-s1234,24000,400,0.31,0.31,false,cache_mask_modes_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_modes_v1/production/corrected_validation_histories.csv,4bcc3045afde9ab91db62c6445d0660923ac663e18d314a138fecdbecf1fba9d,4e54078f7b714de634c7c36dc29bd3e3f41ad945672c82d83d63052cd662b3d8,PASS | |
| qwen,by_representation/full_hidden_sequence/prefix_components/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_prefix_components_hidden_sequence_t8_final_prefix_direct_only_seed1234,qwen-prefix-direct-only-t8-s1234,25000,400,0.3125,0.3125,true,cache_mask_modes_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_modes_v1/production/corrected_validation_histories.csv,0f503cc69229995c567f68e7e8ac0e47b02291392848820dec53ad65aebd3fcd,81088d63a03ca3437c242a2af5ecf1dbb43ad0bac6eb0267636d6c5b5960b24e,PASS | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,1000,400,0,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,2000,400,0,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,3000,400,0.005,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,4000,400,0.0075,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,5000,400,0.0075,,true,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,6000,400,0,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,7000,400,0,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,8000,400,0,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,9000,400,0,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,10000,400,0,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,11000,400,0,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,12000,400,0,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,13000,400,0,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,14000,400,0,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,15000,400,0,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,16000,400,0,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,17000,400,0,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,18000,400,0,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,19000,400,0,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,20000,400,0,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,21000,400,0,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,22000,400,0,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,23000,400,0.0025,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,24000,400,0,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/last_token/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_last_token_t8_final_primary_matched_seed1234,qwen-primary-final-t8-s1234,25000,400,0,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/full_hidden_sequence/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_hidden_sequence_t8_final_primary_matched_seed1234,qwen-primary-full-t8-s1234,21000,400,0.2425,0.24,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,7a72f2e467bc1624b5225ca89387d6710ef9c8331b113055f0f3d911109e89c1,, | |
| qwen,by_representation/full_hidden_sequence/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_hidden_sequence_t8_final_primary_matched_seed1234,qwen-primary-full-t8-s1234,22000,400,0.2225,0.22,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,fb624591aa6a6fcbb984c0ff460bf4ca41a423699017fc82f4be41a198f40ca0,, | |
| qwen,by_representation/full_hidden_sequence/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_hidden_sequence_t8_final_primary_matched_seed1234,qwen-primary-full-t8-s1234,23000,400,0.2475,0.245,true,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,486b9459d996223ff4c4f70bd4eb763cb022ae1067956c8934316818394998b4,, | |
| qwen,by_representation/full_hidden_sequence/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_hidden_sequence_t8_final_primary_matched_seed1234,qwen-primary-full-t8-s1234,24000,400,0.2325,0.2325,false,cache_mask_v1,val_real_full,target_prefix_exact_match,later_step,run_docs/cache_mask_v1/current_registry/checkpoint_selection_metric_audit_610.csv,98e76ec32e83e1f8365dd22c2adb3ecf8bdb6b468847c838fc4a07683dbc289a,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,1000,400,0.005,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,2000,400,0.005,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,3000,400,0.0075,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,4000,400,0.0075,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,5000,400,0.0025,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,6000,400,0.01,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,7000,400,0.0175,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,8000,400,0.0175,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,9000,400,0.02,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,10000,400,0.0275,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,11000,400,0.0275,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,12000,400,0.0275,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,13000,400,0.0325,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,14000,400,0.0325,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,15000,400,0.0325,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,16000,400,0.0275,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,17000,400,0.0375,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,18000,400,0.0325,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,19000,400,0.04,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,20000,400,0.04,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,21000,400,0.0325,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,22000,400,0.0375,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,23000,400,0.04,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,24000,400,0.0475,,true,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |
| qwen,by_representation/pooled/main_t8/train/qwen2p5_0p5b_eosfix_v2/qwen2p5_0p5b_eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,qwen-primary-pooled-t8-s1234,25000,400,0.04,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Qwen2.5-0.5B-Instruct/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,, | |