| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.27705947543405984, | |
| "eval_steps": 100, | |
| "global_step": 1500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01847063169560399, | |
| "grad_norm": 17.696088790893555, | |
| "learning_rate": 1.980139427847242e-05, | |
| "loss": 0.4835, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.01847063169560399, | |
| "eval_cosine_accuracy@1": 0.8868686868686869, | |
| "eval_cosine_accuracy@10": 0.9579124579124579, | |
| "eval_cosine_accuracy@3": 0.9343434343434344, | |
| "eval_cosine_accuracy@5": 0.9454545454545454, | |
| "eval_cosine_map@100": 0.9137674463447905, | |
| "eval_cosine_mrr@10": 0.9124690021912245, | |
| "eval_cosine_ndcg@10": 0.9236137131767355, | |
| "eval_cosine_precision@1": 0.8868686868686869, | |
| "eval_cosine_precision@10": 0.09579124579124577, | |
| "eval_cosine_precision@3": 0.3114478114478115, | |
| "eval_cosine_precision@5": 0.18909090909090906, | |
| "eval_cosine_recall@1": 0.8868686868686869, | |
| "eval_cosine_recall@10": 0.9579124579124579, | |
| "eval_cosine_recall@3": 0.9343434343434344, | |
| "eval_cosine_recall@5": 0.9454545454545454, | |
| "eval_loss": 0.07506837695837021, | |
| "eval_runtime": 49.7303, | |
| "eval_samples_per_second": 140.759, | |
| "eval_sequential_score": 0.9137674463447905, | |
| "eval_steps_per_second": 8.808, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03694126339120798, | |
| "grad_norm": 2.3271963596343994, | |
| "learning_rate": 1.917211301505453e-05, | |
| "loss": 0.0646, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.03694126339120798, | |
| "eval_cosine_accuracy@1": 0.9195286195286195, | |
| "eval_cosine_accuracy@10": 0.967003367003367, | |
| "eval_cosine_accuracy@3": 0.9518518518518518, | |
| "eval_cosine_accuracy@5": 0.9612794612794613, | |
| "eval_cosine_map@100": 0.9383862806127067, | |
| "eval_cosine_mrr@10": 0.937248810859922, | |
| "eval_cosine_ndcg@10": 0.9445895366693552, | |
| "eval_cosine_precision@1": 0.9195286195286195, | |
| "eval_cosine_precision@10": 0.0967003367003367, | |
| "eval_cosine_precision@3": 0.317283950617284, | |
| "eval_cosine_precision@5": 0.19225589225589226, | |
| "eval_cosine_recall@1": 0.9195286195286195, | |
| "eval_cosine_recall@10": 0.967003367003367, | |
| "eval_cosine_recall@3": 0.9518518518518518, | |
| "eval_cosine_recall@5": 0.9612794612794613, | |
| "eval_loss": 0.05896875262260437, | |
| "eval_runtime": 51.3489, | |
| "eval_samples_per_second": 136.322, | |
| "eval_sequential_score": 0.9383862806127067, | |
| "eval_steps_per_second": 8.53, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.05541189508681197, | |
| "grad_norm": 3.7376925945281982, | |
| "learning_rate": 1.8139290433532415e-05, | |
| "loss": 0.0594, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.05541189508681197, | |
| "eval_cosine_accuracy@1": 0.9303030303030303, | |
| "eval_cosine_accuracy@10": 0.9737373737373738, | |
| "eval_cosine_accuracy@3": 0.9579124579124579, | |
| "eval_cosine_accuracy@5": 0.9656565656565657, | |
| "eval_cosine_map@100": 0.9462389210939364, | |
| "eval_cosine_mrr@10": 0.9454570840681954, | |
| "eval_cosine_ndcg@10": 0.9523521034308455, | |
| "eval_cosine_precision@1": 0.9303030303030303, | |
| "eval_cosine_precision@10": 0.09737373737373735, | |
| "eval_cosine_precision@3": 0.31930415263748596, | |
| "eval_cosine_precision@5": 0.1931313131313131, | |
| "eval_cosine_recall@1": 0.9303030303030303, | |
| "eval_cosine_recall@10": 0.9737373737373738, | |
| "eval_cosine_recall@3": 0.9579124579124579, | |
| "eval_cosine_recall@5": 0.9656565656565657, | |
| "eval_loss": 0.051894593983888626, | |
| "eval_runtime": 49.496, | |
| "eval_samples_per_second": 141.426, | |
| "eval_sequential_score": 0.9462389210939364, | |
| "eval_steps_per_second": 8.849, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.07388252678241596, | |
| "grad_norm": 0.3877984583377838, | |
| "learning_rate": 1.6748367163042577e-05, | |
| "loss": 0.0471, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.07388252678241596, | |
| "eval_cosine_accuracy@1": 0.9367003367003367, | |
| "eval_cosine_accuracy@10": 0.9750841750841751, | |
| "eval_cosine_accuracy@3": 0.9612794612794613, | |
| "eval_cosine_accuracy@5": 0.969023569023569, | |
| "eval_cosine_map@100": 0.9513700816079773, | |
| "eval_cosine_mrr@10": 0.9505351130351131, | |
| "eval_cosine_ndcg@10": 0.9565510675566292, | |
| "eval_cosine_precision@1": 0.9367003367003367, | |
| "eval_cosine_precision@10": 0.09750841750841752, | |
| "eval_cosine_precision@3": 0.3204264870931538, | |
| "eval_cosine_precision@5": 0.19380471380471379, | |
| "eval_cosine_recall@1": 0.9367003367003367, | |
| "eval_cosine_recall@10": 0.9750841750841751, | |
| "eval_cosine_recall@3": 0.9612794612794613, | |
| "eval_cosine_recall@5": 0.969023569023569, | |
| "eval_loss": 0.04832224175333977, | |
| "eval_runtime": 49.2695, | |
| "eval_samples_per_second": 142.076, | |
| "eval_sequential_score": 0.9513700816079773, | |
| "eval_steps_per_second": 8.89, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.09235315847801995, | |
| "grad_norm": 0.9424126744270325, | |
| "learning_rate": 1.5060539027168317e-05, | |
| "loss": 0.0524, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09235315847801995, | |
| "eval_cosine_accuracy@1": 0.9387205387205387, | |
| "eval_cosine_accuracy@10": 0.9787878787878788, | |
| "eval_cosine_accuracy@3": 0.9622895622895623, | |
| "eval_cosine_accuracy@5": 0.9703703703703703, | |
| "eval_cosine_map@100": 0.9530933506069861, | |
| "eval_cosine_mrr@10": 0.9525124258457593, | |
| "eval_cosine_ndcg@10": 0.9588799906525647, | |
| "eval_cosine_precision@1": 0.9387205387205387, | |
| "eval_cosine_precision@10": 0.09787878787878787, | |
| "eval_cosine_precision@3": 0.3207631874298541, | |
| "eval_cosine_precision@5": 0.19407407407407404, | |
| "eval_cosine_recall@1": 0.9387205387205387, | |
| "eval_cosine_recall@10": 0.9787878787878788, | |
| "eval_cosine_recall@3": 0.9622895622895623, | |
| "eval_cosine_recall@5": 0.9703703703703703, | |
| "eval_loss": 0.04548173025250435, | |
| "eval_runtime": 49.8784, | |
| "eval_samples_per_second": 140.341, | |
| "eval_sequential_score": 0.9530933506069861, | |
| "eval_steps_per_second": 8.781, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.11082379017362394, | |
| "grad_norm": 1.4559208154678345, | |
| "learning_rate": 1.315006463889948e-05, | |
| "loss": 0.0435, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.11082379017362394, | |
| "eval_cosine_accuracy@1": 0.94006734006734, | |
| "eval_cosine_accuracy@10": 0.9784511784511785, | |
| "eval_cosine_accuracy@3": 0.965993265993266, | |
| "eval_cosine_accuracy@5": 0.9713804713804713, | |
| "eval_cosine_map@100": 0.9545577685553948, | |
| "eval_cosine_mrr@10": 0.953849339960451, | |
| "eval_cosine_ndcg@10": 0.9598872198067653, | |
| "eval_cosine_precision@1": 0.94006734006734, | |
| "eval_cosine_precision@10": 0.09784511784511785, | |
| "eval_cosine_precision@3": 0.32199775533108865, | |
| "eval_cosine_precision@5": 0.19427609427609427, | |
| "eval_cosine_recall@1": 0.94006734006734, | |
| "eval_cosine_recall@10": 0.9784511784511785, | |
| "eval_cosine_recall@3": 0.965993265993266, | |
| "eval_cosine_recall@5": 0.9713804713804713, | |
| "eval_loss": 0.03971650078892708, | |
| "eval_runtime": 50.0185, | |
| "eval_samples_per_second": 139.948, | |
| "eval_sequential_score": 0.9545577685553948, | |
| "eval_steps_per_second": 8.757, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.12929442186922793, | |
| "grad_norm": 4.4629669189453125, | |
| "learning_rate": 1.1100998277940316e-05, | |
| "loss": 0.0336, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.12929442186922793, | |
| "eval_cosine_accuracy@1": 0.9404040404040404, | |
| "eval_cosine_accuracy@10": 0.9804713804713805, | |
| "eval_cosine_accuracy@3": 0.9646464646464646, | |
| "eval_cosine_accuracy@5": 0.9717171717171718, | |
| "eval_cosine_map@100": 0.9549470882860054, | |
| "eval_cosine_mrr@10": 0.9544308160974829, | |
| "eval_cosine_ndcg@10": 0.9607541640834583, | |
| "eval_cosine_precision@1": 0.9404040404040404, | |
| "eval_cosine_precision@10": 0.09804713804713802, | |
| "eval_cosine_precision@3": 0.32154882154882153, | |
| "eval_cosine_precision@5": 0.19434343434343435, | |
| "eval_cosine_recall@1": 0.9404040404040404, | |
| "eval_cosine_recall@10": 0.9804713804713805, | |
| "eval_cosine_recall@3": 0.9646464646464646, | |
| "eval_cosine_recall@5": 0.9717171717171718, | |
| "eval_loss": 0.03936752304434776, | |
| "eval_runtime": 49.3604, | |
| "eval_samples_per_second": 141.814, | |
| "eval_sequential_score": 0.9549470882860054, | |
| "eval_steps_per_second": 8.874, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.1477650535648319, | |
| "grad_norm": 18.81490135192871, | |
| "learning_rate": 9.003491792488438e-06, | |
| "loss": 0.0344, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.1477650535648319, | |
| "eval_cosine_accuracy@1": 0.9424242424242424, | |
| "eval_cosine_accuracy@10": 0.9791245791245792, | |
| "eval_cosine_accuracy@3": 0.9673400673400674, | |
| "eval_cosine_accuracy@5": 0.9730639730639731, | |
| "eval_cosine_map@100": 0.9565285707123049, | |
| "eval_cosine_mrr@10": 0.9558751536529315, | |
| "eval_cosine_ndcg@10": 0.9615848253233655, | |
| "eval_cosine_precision@1": 0.9424242424242424, | |
| "eval_cosine_precision@10": 0.0979124579124579, | |
| "eval_cosine_precision@3": 0.32244668911335583, | |
| "eval_cosine_precision@5": 0.19461279461279463, | |
| "eval_cosine_recall@1": 0.9424242424242424, | |
| "eval_cosine_recall@10": 0.9791245791245792, | |
| "eval_cosine_recall@3": 0.9673400673400674, | |
| "eval_cosine_recall@5": 0.9730639730639731, | |
| "eval_loss": 0.03738004341721535, | |
| "eval_runtime": 50.7853, | |
| "eval_samples_per_second": 137.835, | |
| "eval_sequential_score": 0.9565285707123049, | |
| "eval_steps_per_second": 8.625, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.16623568526043592, | |
| "grad_norm": 40.96598434448242, | |
| "learning_rate": 6.9498282290438235e-06, | |
| "loss": 0.0393, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.16623568526043592, | |
| "eval_cosine_accuracy@1": 0.9430976430976431, | |
| "eval_cosine_accuracy@10": 0.9801346801346801, | |
| "eval_cosine_accuracy@3": 0.9666666666666667, | |
| "eval_cosine_accuracy@5": 0.9723905723905724, | |
| "eval_cosine_map@100": 0.9567946373289521, | |
| "eval_cosine_mrr@10": 0.9561974239752019, | |
| "eval_cosine_ndcg@10": 0.9620315668852821, | |
| "eval_cosine_precision@1": 0.9430976430976431, | |
| "eval_cosine_precision@10": 0.098013468013468, | |
| "eval_cosine_precision@3": 0.32222222222222224, | |
| "eval_cosine_precision@5": 0.19447811447811444, | |
| "eval_cosine_recall@1": 0.9430976430976431, | |
| "eval_cosine_recall@10": 0.9801346801346801, | |
| "eval_cosine_recall@3": 0.9666666666666667, | |
| "eval_cosine_recall@5": 0.9723905723905724, | |
| "eval_loss": 0.036062091588974, | |
| "eval_runtime": 50.1457, | |
| "eval_samples_per_second": 139.593, | |
| "eval_sequential_score": 0.9567946373289521, | |
| "eval_steps_per_second": 8.735, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.1847063169560399, | |
| "grad_norm": 7.633666038513184, | |
| "learning_rate": 5.030361696847706e-06, | |
| "loss": 0.0451, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1847063169560399, | |
| "eval_cosine_accuracy@1": 0.9447811447811448, | |
| "eval_cosine_accuracy@10": 0.9804713804713805, | |
| "eval_cosine_accuracy@3": 0.9673400673400674, | |
| "eval_cosine_accuracy@5": 0.9720538720538721, | |
| "eval_cosine_map@100": 0.9577987764578036, | |
| "eval_cosine_mrr@10": 0.9572219549997326, | |
| "eval_cosine_ndcg@10": 0.9628692157043424, | |
| "eval_cosine_precision@1": 0.9447811447811448, | |
| "eval_cosine_precision@10": 0.09804713804713805, | |
| "eval_cosine_precision@3": 0.32244668911335583, | |
| "eval_cosine_precision@5": 0.19441077441077437, | |
| "eval_cosine_recall@1": 0.9447811447811448, | |
| "eval_cosine_recall@10": 0.9804713804713805, | |
| "eval_cosine_recall@3": 0.9673400673400674, | |
| "eval_cosine_recall@5": 0.9720538720538721, | |
| "eval_loss": 0.03610473498702049, | |
| "eval_runtime": 49.6014, | |
| "eval_samples_per_second": 141.125, | |
| "eval_sequential_score": 0.9577987764578036, | |
| "eval_steps_per_second": 8.83, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.20317694865164387, | |
| "grad_norm": 4.241499900817871, | |
| "learning_rate": 3.329542098903674e-06, | |
| "loss": 0.0278, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.20317694865164387, | |
| "eval_cosine_accuracy@1": 0.9430976430976431, | |
| "eval_cosine_accuracy@10": 0.9804713804713805, | |
| "eval_cosine_accuracy@3": 0.9666666666666667, | |
| "eval_cosine_accuracy@5": 0.9717171717171718, | |
| "eval_cosine_map@100": 0.9568101043018846, | |
| "eval_cosine_mrr@10": 0.9562193362193362, | |
| "eval_cosine_ndcg@10": 0.9621111011663721, | |
| "eval_cosine_precision@1": 0.9430976430976431, | |
| "eval_cosine_precision@10": 0.09804713804713805, | |
| "eval_cosine_precision@3": 0.32222222222222224, | |
| "eval_cosine_precision@5": 0.19434343434343435, | |
| "eval_cosine_recall@1": 0.9430976430976431, | |
| "eval_cosine_recall@10": 0.9804713804713805, | |
| "eval_cosine_recall@3": 0.9666666666666667, | |
| "eval_cosine_recall@5": 0.9717171717171718, | |
| "eval_loss": 0.035753391683101654, | |
| "eval_runtime": 50.7364, | |
| "eval_samples_per_second": 137.968, | |
| "eval_sequential_score": 0.9568101043018846, | |
| "eval_steps_per_second": 8.633, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.22164758034724788, | |
| "grad_norm": 10.087759971618652, | |
| "learning_rate": 1.9221996276968523e-06, | |
| "loss": 0.0332, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.22164758034724788, | |
| "eval_cosine_accuracy@1": 0.9437710437710438, | |
| "eval_cosine_accuracy@10": 0.9801346801346801, | |
| "eval_cosine_accuracy@3": 0.9666666666666667, | |
| "eval_cosine_accuracy@5": 0.9720538720538721, | |
| "eval_cosine_map@100": 0.9572367070457485, | |
| "eval_cosine_mrr@10": 0.9566237774571107, | |
| "eval_cosine_ndcg@10": 0.9623501597917694, | |
| "eval_cosine_precision@1": 0.9437710437710438, | |
| "eval_cosine_precision@10": 0.09801346801346802, | |
| "eval_cosine_precision@3": 0.32222222222222224, | |
| "eval_cosine_precision@5": 0.19441077441077437, | |
| "eval_cosine_recall@1": 0.9437710437710438, | |
| "eval_cosine_recall@10": 0.9801346801346801, | |
| "eval_cosine_recall@3": 0.9666666666666667, | |
| "eval_cosine_recall@5": 0.9720538720538721, | |
| "eval_loss": 0.03558042645454407, | |
| "eval_runtime": 50.0835, | |
| "eval_samples_per_second": 139.767, | |
| "eval_sequential_score": 0.9572367070457485, | |
| "eval_steps_per_second": 8.745, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.24011821204285186, | |
| "grad_norm": 17.43052864074707, | |
| "learning_rate": 8.702524949765645e-07, | |
| "loss": 0.0317, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.24011821204285186, | |
| "eval_cosine_accuracy@1": 0.9441077441077441, | |
| "eval_cosine_accuracy@10": 0.9797979797979798, | |
| "eval_cosine_accuracy@3": 0.9673400673400674, | |
| "eval_cosine_accuracy@5": 0.9720538720538721, | |
| "eval_cosine_map@100": 0.9574821374909939, | |
| "eval_cosine_mrr@10": 0.9568467746245525, | |
| "eval_cosine_ndcg@10": 0.9624426218971004, | |
| "eval_cosine_precision@1": 0.9441077441077441, | |
| "eval_cosine_precision@10": 0.09797979797979799, | |
| "eval_cosine_precision@3": 0.32244668911335583, | |
| "eval_cosine_precision@5": 0.19441077441077437, | |
| "eval_cosine_recall@1": 0.9441077441077441, | |
| "eval_cosine_recall@10": 0.9797979797979798, | |
| "eval_cosine_recall@3": 0.9673400673400674, | |
| "eval_cosine_recall@5": 0.9720538720538721, | |
| "eval_loss": 0.03541325777769089, | |
| "eval_runtime": 49.4999, | |
| "eval_samples_per_second": 141.414, | |
| "eval_sequential_score": 0.9574821374909939, | |
| "eval_steps_per_second": 8.848, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.25858884373845586, | |
| "grad_norm": 0.8455698490142822, | |
| "learning_rate": 2.199827441298863e-07, | |
| "loss": 0.026, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.25858884373845586, | |
| "eval_cosine_accuracy@1": 0.9441077441077441, | |
| "eval_cosine_accuracy@10": 0.9797979797979798, | |
| "eval_cosine_accuracy@3": 0.9673400673400674, | |
| "eval_cosine_accuracy@5": 0.9720538720538721, | |
| "eval_cosine_map@100": 0.9574340839544937, | |
| "eval_cosine_mrr@10": 0.9567861151194483, | |
| "eval_cosine_ndcg@10": 0.9624031906320126, | |
| "eval_cosine_precision@1": 0.9441077441077441, | |
| "eval_cosine_precision@10": 0.09797979797979799, | |
| "eval_cosine_precision@3": 0.32244668911335583, | |
| "eval_cosine_precision@5": 0.19441077441077437, | |
| "eval_cosine_recall@1": 0.9441077441077441, | |
| "eval_cosine_recall@10": 0.9797979797979798, | |
| "eval_cosine_recall@3": 0.9673400673400674, | |
| "eval_cosine_recall@5": 0.9720538720538721, | |
| "eval_loss": 0.035498470067977905, | |
| "eval_runtime": 50.8752, | |
| "eval_samples_per_second": 137.592, | |
| "eval_sequential_score": 0.9574340839544937, | |
| "eval_steps_per_second": 8.609, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.27705947543405984, | |
| "grad_norm": 3.3353021144866943, | |
| "learning_rate": 0.0, | |
| "loss": 0.0442, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.27705947543405984, | |
| "eval_cosine_accuracy@1": 0.9437710437710438, | |
| "eval_cosine_accuracy@10": 0.9801346801346801, | |
| "eval_cosine_accuracy@3": 0.967003367003367, | |
| "eval_cosine_accuracy@5": 0.9723905723905724, | |
| "eval_cosine_map@100": 0.9572829070357247, | |
| "eval_cosine_mrr@10": 0.9566718775052107, | |
| "eval_cosine_ndcg@10": 0.9623908732460177, | |
| "eval_cosine_precision@1": 0.9437710437710438, | |
| "eval_cosine_precision@10": 0.09801346801346802, | |
| "eval_cosine_precision@3": 0.322334455667789, | |
| "eval_cosine_precision@5": 0.19447811447811444, | |
| "eval_cosine_recall@1": 0.9437710437710438, | |
| "eval_cosine_recall@10": 0.9801346801346801, | |
| "eval_cosine_recall@3": 0.967003367003367, | |
| "eval_cosine_recall@5": 0.9723905723905724, | |
| "eval_loss": 0.035526029765605927, | |
| "eval_runtime": 45.9521, | |
| "eval_samples_per_second": 152.333, | |
| "eval_sequential_score": 0.9572829070357247, | |
| "eval_steps_per_second": 9.532, | |
| "step": 1500 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 1500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |