| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0, |
| "eval_steps": 79, |
| "global_step": 782, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0, |
| "eval_cosine_accuracy@1": 0.232, |
| "eval_cosine_accuracy@10": 0.456, |
| "eval_cosine_accuracy@3": 0.3365, |
| "eval_cosine_accuracy@5": 0.389, |
| "eval_cosine_map@100": 0.14334533833340427, |
| "eval_cosine_mrr@10": 0.2988200396825396, |
| "eval_cosine_ndcg@10": 0.19590710853524823, |
| "eval_cosine_precision@1": 0.232, |
| "eval_cosine_precision@10": 0.10355000000000002, |
| "eval_cosine_precision@3": 0.177, |
| "eval_cosine_precision@5": 0.1452, |
| "eval_cosine_recall@1": 0.0548117388039026, |
| "eval_cosine_recall@10": 0.18984275866485092, |
| "eval_cosine_recall@3": 0.11065666057433865, |
| "eval_cosine_recall@5": 0.1464970398055657, |
| "eval_loss": 2.950530767440796, |
| "eval_runtime": 28.7759, |
| "eval_samples_per_second": 69.502, |
| "eval_steps_per_second": 1.112, |
| "step": 0 |
| }, |
| { |
| "epoch": 0.10112, |
| "grad_norm": 4.060101509094238, |
| "learning_rate": 4.936708860759494e-05, |
| "loss": 1.7256, |
| "step": 79 |
| }, |
| { |
| "epoch": 0.10112, |
| "eval_cosine_accuracy@1": 0.4915, |
| "eval_cosine_accuracy@10": 0.709, |
| "eval_cosine_accuracy@3": 0.6055, |
| "eval_cosine_accuracy@5": 0.65, |
| "eval_cosine_map@100": 0.37501954096960194, |
| "eval_cosine_mrr@10": 0.5594632936507937, |
| "eval_cosine_ndcg@10": 0.4471963096331387, |
| "eval_cosine_precision@1": 0.4915, |
| "eval_cosine_precision@10": 0.20455, |
| "eval_cosine_precision@3": 0.3835, |
| "eval_cosine_precision@5": 0.30490000000000006, |
| "eval_cosine_recall@1": 0.15563947633799366, |
| "eval_cosine_recall@10": 0.4411529898745371, |
| "eval_cosine_recall@3": 0.30525816129147315, |
| "eval_cosine_recall@5": 0.3661646057217302, |
| "eval_loss": 0.869914174079895, |
| "eval_runtime": 32.9054, |
| "eval_samples_per_second": 60.78, |
| "eval_steps_per_second": 0.972, |
| "step": 79 |
| }, |
| { |
| "epoch": 0.20224, |
| "grad_norm": 3.455092668533325, |
| "learning_rate": 4.4452347083926033e-05, |
| "loss": 0.8044, |
| "step": 158 |
| }, |
| { |
| "epoch": 0.20224, |
| "eval_cosine_accuracy@1": 0.538, |
| "eval_cosine_accuracy@10": 0.7835, |
| "eval_cosine_accuracy@3": 0.671, |
| "eval_cosine_accuracy@5": 0.72, |
| "eval_cosine_map@100": 0.4288531840469143, |
| "eval_cosine_mrr@10": 0.6170097222222221, |
| "eval_cosine_ndcg@10": 0.5071737492167014, |
| "eval_cosine_precision@1": 0.538, |
| "eval_cosine_precision@10": 0.22805, |
| "eval_cosine_precision@3": 0.419, |
| "eval_cosine_precision@5": 0.336, |
| "eval_cosine_recall@1": 0.18509011905489597, |
| "eval_cosine_recall@10": 0.5088450856238615, |
| "eval_cosine_recall@3": 0.35023499358379173, |
| "eval_cosine_recall@5": 0.42149330607177377, |
| "eval_loss": 0.6144608855247498, |
| "eval_runtime": 33.0639, |
| "eval_samples_per_second": 60.489, |
| "eval_steps_per_second": 0.968, |
| "step": 158 |
| }, |
| { |
| "epoch": 0.30336, |
| "grad_norm": 3.299586534500122, |
| "learning_rate": 3.883357041251778e-05, |
| "loss": 0.6254, |
| "step": 237 |
| }, |
| { |
| "epoch": 0.30336, |
| "eval_cosine_accuracy@1": 0.571, |
| "eval_cosine_accuracy@10": 0.807, |
| "eval_cosine_accuracy@3": 0.697, |
| "eval_cosine_accuracy@5": 0.747, |
| "eval_cosine_map@100": 0.45768471305297626, |
| "eval_cosine_mrr@10": 0.6465369047619048, |
| "eval_cosine_ndcg@10": 0.5352649581411899, |
| "eval_cosine_precision@1": 0.571, |
| "eval_cosine_precision@10": 0.23940000000000003, |
| "eval_cosine_precision@3": 0.446, |
| "eval_cosine_precision@5": 0.3534000000000001, |
| "eval_cosine_recall@1": 0.19728028074863738, |
| "eval_cosine_recall@10": 0.5336122046159524, |
| "eval_cosine_recall@3": 0.3745295533176793, |
| "eval_cosine_recall@5": 0.44429268806860206, |
| "eval_loss": 0.5068339705467224, |
| "eval_runtime": 32.9503, |
| "eval_samples_per_second": 60.697, |
| "eval_steps_per_second": 0.971, |
| "step": 237 |
| }, |
| { |
| "epoch": 0.40448, |
| "grad_norm": 2.563180685043335, |
| "learning_rate": 3.321479374110953e-05, |
| "loss": 0.5215, |
| "step": 316 |
| }, |
| { |
| "epoch": 0.40448, |
| "eval_cosine_accuracy@1": 0.6075, |
| "eval_cosine_accuracy@10": 0.855, |
| "eval_cosine_accuracy@3": 0.7495, |
| "eval_cosine_accuracy@5": 0.8005, |
| "eval_cosine_map@100": 0.4886299583708059, |
| "eval_cosine_mrr@10": 0.6879220238095239, |
| "eval_cosine_ndcg@10": 0.5725427973832125, |
| "eval_cosine_precision@1": 0.6075, |
| "eval_cosine_precision@10": 0.25585, |
| "eval_cosine_precision@3": 0.4696666666666666, |
| "eval_cosine_precision@5": 0.3763000000000001, |
| "eval_cosine_recall@1": 0.21129895871002893, |
| "eval_cosine_recall@10": 0.5768064108469603, |
| "eval_cosine_recall@3": 0.3986053442055735, |
| "eval_cosine_recall@5": 0.4793458893809688, |
| "eval_loss": 0.4230406582355499, |
| "eval_runtime": 32.9072, |
| "eval_samples_per_second": 60.777, |
| "eval_steps_per_second": 0.972, |
| "step": 316 |
| }, |
| { |
| "epoch": 0.5056, |
| "grad_norm": 2.914031744003296, |
| "learning_rate": 2.759601706970128e-05, |
| "loss": 0.4507, |
| "step": 395 |
| }, |
| { |
| "epoch": 0.5056, |
| "eval_cosine_accuracy@1": 0.6285, |
| "eval_cosine_accuracy@10": 0.8675, |
| "eval_cosine_accuracy@3": 0.753, |
| "eval_cosine_accuracy@5": 0.812, |
| "eval_cosine_map@100": 0.5064669285677783, |
| "eval_cosine_mrr@10": 0.7049886904761903, |
| "eval_cosine_ndcg@10": 0.5906093986384608, |
| "eval_cosine_precision@1": 0.6285, |
| "eval_cosine_precision@10": 0.26185, |
| "eval_cosine_precision@3": 0.4774999999999999, |
| "eval_cosine_precision@5": 0.38320000000000004, |
| "eval_cosine_recall@1": 0.22386914380224288, |
| "eval_cosine_recall@10": 0.5924252077258059, |
| "eval_cosine_recall@3": 0.4076045673568921, |
| "eval_cosine_recall@5": 0.4945358123439019, |
| "eval_loss": 0.3786991834640503, |
| "eval_runtime": 32.9486, |
| "eval_samples_per_second": 60.701, |
| "eval_steps_per_second": 0.971, |
| "step": 395 |
| }, |
| { |
| "epoch": 0.60672, |
| "grad_norm": 2.8829078674316406, |
| "learning_rate": 2.197724039829303e-05, |
| "loss": 0.4014, |
| "step": 474 |
| }, |
| { |
| "epoch": 0.60672, |
| "eval_cosine_accuracy@1": 0.6365, |
| "eval_cosine_accuracy@10": 0.876, |
| "eval_cosine_accuracy@3": 0.771, |
| "eval_cosine_accuracy@5": 0.8255, |
| "eval_cosine_map@100": 0.5197156334384408, |
| "eval_cosine_mrr@10": 0.7141259920634916, |
| "eval_cosine_ndcg@10": 0.601418022062486, |
| "eval_cosine_precision@1": 0.6365, |
| "eval_cosine_precision@10": 0.26645, |
| "eval_cosine_precision@3": 0.48833333333333334, |
| "eval_cosine_precision@5": 0.3937, |
| "eval_cosine_recall@1": 0.22605882194620797, |
| "eval_cosine_recall@10": 0.6025888884821827, |
| "eval_cosine_recall@3": 0.42357208114816375, |
| "eval_cosine_recall@5": 0.509917211803928, |
| "eval_loss": 0.3428954482078552, |
| "eval_runtime": 33.015, |
| "eval_samples_per_second": 60.579, |
| "eval_steps_per_second": 0.969, |
| "step": 474 |
| }, |
| { |
| "epoch": 0.70784, |
| "grad_norm": 2.7049639225006104, |
| "learning_rate": 1.6358463726884778e-05, |
| "loss": 0.3779, |
| "step": 553 |
| }, |
| { |
| "epoch": 0.70784, |
| "eval_cosine_accuracy@1": 0.6455, |
| "eval_cosine_accuracy@10": 0.8815, |
| "eval_cosine_accuracy@3": 0.784, |
| "eval_cosine_accuracy@5": 0.8345, |
| "eval_cosine_map@100": 0.5322933983250332, |
| "eval_cosine_mrr@10": 0.7239722222222216, |
| "eval_cosine_ndcg@10": 0.6129085102767301, |
| "eval_cosine_precision@1": 0.6455, |
| "eval_cosine_precision@10": 0.27020000000000005, |
| "eval_cosine_precision@3": 0.49716666666666665, |
| "eval_cosine_precision@5": 0.39830000000000004, |
| "eval_cosine_recall@1": 0.2337385415837256, |
| "eval_cosine_recall@10": 0.6114790331062327, |
| "eval_cosine_recall@3": 0.43581210236649, |
| "eval_cosine_recall@5": 0.5201249831864138, |
| "eval_loss": 0.32787367701530457, |
| "eval_runtime": 33.0268, |
| "eval_samples_per_second": 60.557, |
| "eval_steps_per_second": 0.969, |
| "step": 553 |
| }, |
| { |
| "epoch": 0.80896, |
| "grad_norm": 2.0675265789031982, |
| "learning_rate": 1.073968705547653e-05, |
| "loss": 0.3463, |
| "step": 632 |
| }, |
| { |
| "epoch": 0.80896, |
| "eval_cosine_accuracy@1": 0.656, |
| "eval_cosine_accuracy@10": 0.889, |
| "eval_cosine_accuracy@3": 0.793, |
| "eval_cosine_accuracy@5": 0.839, |
| "eval_cosine_map@100": 0.5425920958735082, |
| "eval_cosine_mrr@10": 0.7340821428571421, |
| "eval_cosine_ndcg@10": 0.6223116530478937, |
| "eval_cosine_precision@1": 0.656, |
| "eval_cosine_precision@10": 0.27415, |
| "eval_cosine_precision@3": 0.504, |
| "eval_cosine_precision@5": 0.4019000000000001, |
| "eval_cosine_recall@1": 0.23843478790276604, |
| "eval_cosine_recall@10": 0.6201721618621493, |
| "eval_cosine_recall@3": 0.442197840531875, |
| "eval_cosine_recall@5": 0.5237564546607093, |
| "eval_loss": 0.30987244844436646, |
| "eval_runtime": 33.1417, |
| "eval_samples_per_second": 60.347, |
| "eval_steps_per_second": 0.966, |
| "step": 632 |
| }, |
| { |
| "epoch": 0.91008, |
| "grad_norm": 1.9332996606826782, |
| "learning_rate": 5.120910384068279e-06, |
| "loss": 0.3202, |
| "step": 711 |
| }, |
| { |
| "epoch": 0.91008, |
| "eval_cosine_accuracy@1": 0.6545, |
| "eval_cosine_accuracy@10": 0.89, |
| "eval_cosine_accuracy@3": 0.7925, |
| "eval_cosine_accuracy@5": 0.841, |
| "eval_cosine_map@100": 0.5411107394124216, |
| "eval_cosine_mrr@10": 0.7325833333333327, |
| "eval_cosine_ndcg@10": 0.6220169640997574, |
| "eval_cosine_precision@1": 0.6545, |
| "eval_cosine_precision@10": 0.27425, |
| "eval_cosine_precision@3": 0.5031666666666667, |
| "eval_cosine_precision@5": 0.4011, |
| "eval_cosine_recall@1": 0.23852962917260728, |
| "eval_cosine_recall@10": 0.6212349911293668, |
| "eval_cosine_recall@3": 0.4422107764041644, |
| "eval_cosine_recall@5": 0.5245186089315591, |
| "eval_loss": 0.29744598269462585, |
| "eval_runtime": 33.0411, |
| "eval_samples_per_second": 60.531, |
| "eval_steps_per_second": 0.968, |
| "step": 711 |
| } |
| ], |
| "logging_steps": 79, |
| "max_steps": 782, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 79, |
| "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": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
