vec2seq-llama2-reviewer-checkpoints / metadata /validation_histories.csv
vec2seq-artifact's picture
Add files using upload-large-folder tool
b49ab8d verified
Raw
History Blame Contribute Delete
33.6 kB
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
llama,validation_exact_v1/experiments/capacity_study/train/fixed_linear/seed1234_train_87dcc3ad6872,llama-capacity-fixed-linear-t8-s1234,8000,400,0.7375,0.735,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,f8462a66af25e0e57685d4e48a8c619edfb9cd334a7cdada2d911b22b93a96dc,f2d6afba4aea2dda24663de0852133fa58921c00855237fe313548fbd494990e,PASS
llama,validation_exact_v1/experiments/capacity_study/train/fixed_linear/seed1234_train_87dcc3ad6872,llama-capacity-fixed-linear-t8-s1234,9000,400,0.7375,0.735,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,ecdce001c4b22e5dbd0dd636c62c3b17d79f9828bfb1d60e566415cd55b96fe9,4f1213f4544f1a442228791b994c41a94deaa7a3c7ff29ec247f0eb1e3da1358,PASS
llama,validation_exact_v1/experiments/capacity_study/train/fixed_linear/seed1234_train_87dcc3ad6872,llama-capacity-fixed-linear-t8-s1234,10000,400,0.73,0.7275,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,a02513d8e0a91074baf71187fed20cbead7cbfda5de9db65a093251f4cac9b8a,5a888ebb64ebd463692988982ffe5f07b54da1762416e75ee1fda26c244b3e59,PASS
llama,validation_exact_v1/experiments/capacity_study/train/fixed_linear/seed1234_train_87dcc3ad6872,llama-capacity-fixed-linear-t8-s1234,14000,400,0.715,0.7125,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,11ab4e8df72fdf8954d68304c9213a8d90df4f3e0a6d08f00d85a87cbd8ad1f8,6389b0bd2a5f8415f99d34147500a355c3411ae7a83491f3f1bd7e09a9a8b908,PASS
llama,validation_exact_v1/experiments/capacity_study/train/fixed_residual_mlp/seed1234_train_b571d6924f6f,llama-capacity-fixed-residual-mlp-t8-s1234,10000,400,0.81,0.8075,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,24a2df4604b89ad62dcab138efdd23e5114999c0c119878b371959190f269268,2253b58a7f396547f194cb56acaf753db579339d09f481daa5c9bd439398fce2,PASS
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,1000,400,0.08,0.0025,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,4e69e244b1b1a8bda59e9127a63c3c1186f58569c6f2e1abdd11aebcebaf392f,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,2000,400,0.235,0.0575,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,e3e0fef2bc1134eacbcf641481044d9a7b631519d23319e8141d3455cdc3227f,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,3000,400,0.375,0.105,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,0dbac1f2ada13968bf6835bd4201c0bcb99878ea8bb9da62872ccf13bc6d2d38,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,4000,400,0.4625,0.1325,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,db09773d005e862db053bee79b1177c2a17a0971c3709c0f1c88b51f9b0b2451,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,5000,400,0.56,0.18,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,0a53f0066cb86f75679815f130afc770c9d6978e31bcd46677bbad65302ab0c6,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,6000,400,0.58,0.185,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,28560dd82d5e1880497674f05d66a3c7d990752f33e7a0af44dd67764711b1e8,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,7000,400,0.6625,0.2275,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,3b63ba0d77cf7bbbd4de89a92ff7d52e2c43c2e3a2ca5a1db65f0dffcb9b1d6d,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,8000,400,0.675,0.22,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,2d2062c23d4aae164c4ca5e25da551e5b478575bc231548814242b918bce49fc,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,9000,400,0.6925,0.2275,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,f17d75e01efc75763c655be51c972f5b8312829ce6bcb4fc1685d4b96ce3d722,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,10000,400,0.7025,0.23,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,894d4e98433b57256eaaf794432ec17422b4cc1caa4e0c164149264c4ad43411,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,11000,400,0.6875,0.225,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,be64d567d55ab710576b551aa15b729a44adca9286d15207a84e0439ea098ef9,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,12000,400,0.685,0.2125,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,6f8b82c6e9f2f59068126bac1bcd8388faeeec928f7f7d08f8b219dc3d4b74e2,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,13000,400,0.695,0.2275,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,60430c84d5557b0837e6d0620702bc7b4ecfbf500473a19292042549c8252577,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,14000,400,0.69,0.225,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,5233bb0a84bbc174af071b675ce0103a7c941382016d8dffaf5957e803e23142,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,15000,400,0.68,0.2175,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,e0a5d315f0171edf52432000353cb2bf188c6c32209c6d63baa0887fc01b9059,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,16000,400,0.705,0.23,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,07e6e1d2f8f2550f58f5b3036f9a22c1cd6cba401f10ba44510d0b845cc3dbcb,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,17000,400,0.6825,0.215,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,5476ce07ee2d9713331e021fbaee5e55f80df59dd02904793dec7d58e17a78ac,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,18000,400,0.7075,0.23,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,2edde9bad7e5eb5a02a0bc1d046f9888c1618a2b7257bac369ce29ad6499ff2b,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,19000,400,0.7,0.2175,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,edb1be2868930f459074ba7013c81f9c5a73c7136f666e3728ebdfa1c8c98290,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,20000,400,0.6975,0.22,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,95c1cfc2566c77b0e4196e16a94a38ef6d34b289df06f097fd0aa0f47a2bc509,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,21000,400,0.695,0.215,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,1648c84178cbff4be4f39d9c61fb20cdbbca91b9b5da0a1213e844d6ca8e4273,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,22000,400,0.69,0.22,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,1cb1fcf3ca7772dab7e981f1efc6502f4bd701b040649c4815499513b7e92e20,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,23000,400,0.685,0.22,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,e585603fbe4bdd7bf7e5c5fd75f6d94d5c76c07165ce8fdf0cc21017396de578,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,24000,400,0.705,0.225,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,006f5e49bd731d057ae852bf6b4557daad565a61547357fc4e9e207b4f6b7178,,
llama,by_representation/full_hidden_sequence/length_scaling/train/eosfix_v2/eosfix_v2_length_scaling_hidden_sequence_t32_final_primary_matched_seed1234,llama-length-full-t32-s1234,25000,400,0.6775,0.2125,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,431049d4f7078fe57e91bfaa2087288989aa7ccd0eb683c6c912b829467f1923,,
llama,by_representation/full_hidden_sequence/prefix_components/train/eosfix_v2/eosfix_v2_prefix_components_hidden_sequence_t8_final_prefix_direct_only_seed1234,llama-prefix-direct-only-t8-s1234,11000,400,0.6925,0.69,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,745d6695ee09578b19ae4a81751155f832300a15024c09da207c3e3b8cc3bca9,8aec880ca8b037e6ba84ceaef841f61336c1b850896511020997b211f7073ffe,PASS
llama,by_representation/full_hidden_sequence/prefix_components/train/eosfix_v2/eosfix_v2_prefix_components_hidden_sequence_t8_final_prefix_direct_only_seed1234,llama-prefix-direct-only-t8-s1234,12000,400,0.68,0.6775,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,7a772ac6c9d7ff8040887731b3845b3f768997c0f5ef1fc9c223b775e1d16471,f7624a216dfb74962512dc6b8ad409e64b5fa4b6c41d8853a3f559beba3bdadd,PASS
llama,by_representation/full_hidden_sequence/prefix_components/train/eosfix_v2/eosfix_v2_prefix_components_hidden_sequence_t8_final_prefix_direct_only_seed1234,llama-prefix-direct-only-t8-s1234,13000,400,0.69,0.6875,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,f3fbf5c5b794c953dcd66d0fa8660739ddf8e5395a5cb3aef9f2f7f73f5b20b0,375cf0d96677b9da904cd98653a9b5a727ab36c7d28cee67be6c28418225f3ea,PASS
llama,by_representation/full_hidden_sequence/prefix_components/train/eosfix_v2/eosfix_v2_prefix_components_hidden_sequence_t8_final_prefix_direct_only_seed1234,llama-prefix-direct-only-t8-s1234,14000,400,0.685,0.6825,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,d7023c47f5297da1c9d5e400f087ff862b002abfd3278a06a4aa1403c769bf30,1e790e65a50645b7cc6de4648e1a42ead13b3f8159ed92783003b50ba5922735,PASS
llama,by_representation/full_hidden_sequence/prefix_components/train/eosfix_v2/eosfix_v2_prefix_components_hidden_sequence_t8_final_prefix_direct_only_seed1234,llama-prefix-direct-only-t8-s1234,17000,400,0.6875,0.685,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,a85989de5da27d18cb1b22e411c530620ee9cbb51a2b76090d272ea11722cd84,a54567311340628261d5a045ecd227a6c7ae965f5d688b03258a8616b8f0b928,PASS
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,1000,400,0.0075,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,2000,400,0.0075,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,3000,400,0.01,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,4000,400,0.0175,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,5000,400,0.0275,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,6000,400,0.0225,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,7000,400,0.025,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,8000,400,0.025,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,9000,400,0.0325,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,10000,400,0.0325,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,11000,400,0.035,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,12000,400,0.0325,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,13000,400,0.0325,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,14000,400,0.035,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,15000,400,0.035,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,16000,400,0.0325,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,17000,400,0.035,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,18000,400,0.0325,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,19000,400,0.045,,true,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,20000,400,0.04,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,21000,400,0.0425,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,22000,400,0.04,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,23000,400,0.0425,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,24000,400,0.04,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/last_token/main_t8/train/eosfix_v2/eosfix_v2_primary_last_token_t8_seed1234_hopper96_20260621_031008,llama-primary-final-t8-s1234,25000,400,0.0325,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/full_hidden_sequence/main_t8/train/eosfix_v2/eosfix_v2_main_t8_hidden_sequence_t8_final_primary_matched_seed1234,llama-primary-full-t8-s1234,9000,400,0.7825,0.78,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,6b63f0849c30f8ee5c1a251aed82c43f330af38c157a4e5e6b245d28059254fa,,
llama,by_representation/full_hidden_sequence/main_t8/train/eosfix_v2/eosfix_v2_main_t8_hidden_sequence_t8_final_primary_matched_seed1234,llama-primary-full-t8-s1234,23000,400,0.7925,0.79,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,60464066426b7375cf37327f459906b8884f73679ba80d52f37a3425ad247fb3,,
llama,by_representation/full_hidden_sequence/main_t8/train/eosfix_v2/eosfix_v2_main_t8_hidden_sequence_t8_final_primary_matched_seed1234,llama-primary-full-t8-s1234,24000,400,0.7825,0.78,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,0170bca7a8fa91074ac819849f75dceb26026595da94ffdf700b9123d19bab32,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,1000,400,0.0225,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,2000,400,0.045,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,3000,400,0.08,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,4000,400,0.1125,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,5000,400,0.12,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,6000,400,0.1475,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,7000,400,0.1625,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,8000,400,0.18,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,9000,400,0.215,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,10000,400,0.2325,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,11000,400,0.2075,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,12000,400,0.205,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,13000,400,0.215,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,14000,400,0.21,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,15000,400,0.2525,,true,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,16000,400,0.225,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,17000,400,0.23,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,18000,400,0.2275,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,19000,400,0.2225,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,20000,400,0.2275,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,21000,400,0.235,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,22000,400,0.2075,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,23000,400,0.2225,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,24000,400,0.2325,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,
llama,by_representation/pooled/main_t8/train/eosfix_v2/eosfix_v2_main_t8_pooled_vector_t8_final_primary_matched_seed1234,llama-primary-pooled-t8-s1234,25000,400,0.2225,,false,validation_exact_v1,val_real_full,target_prefix_exact_match,later_step,Llama-2-7b-hf/experiments/reports/cache_mask_v1/historical_validation_histories.csv,,,