| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 4.0, |
| "eval_steps": 100, |
| "global_step": 5048, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.07923930269413629, |
| "grad_norm": 0.36957958340644836, |
| "learning_rate": 9.801980198019804e-06, |
| "loss": 1.0261, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.07923930269413629, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9996037483215332, |
| "eval_loss": 0.009768067859113216, |
| "eval_runtime": 34.4825, |
| "eval_samples_per_second": 146.364, |
| "eval_steps_per_second": 4.582, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.15847860538827258, |
| "grad_norm": 0.15150195360183716, |
| "learning_rate": 1.9702970297029703e-05, |
| "loss": 0.0086, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.15847860538827258, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.999405562877655, |
| "eval_loss": 0.006445930805057287, |
| "eval_runtime": 33.42, |
| "eval_samples_per_second": 151.017, |
| "eval_steps_per_second": 4.728, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.23771790808240886, |
| "grad_norm": 0.24285651743412018, |
| "learning_rate": 2.9603960396039603e-05, |
| "loss": 0.0043, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.23771790808240886, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9996037483215332, |
| "eval_loss": 0.00557651836425066, |
| "eval_runtime": 32.8039, |
| "eval_samples_per_second": 153.854, |
| "eval_steps_per_second": 4.816, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.31695721077654515, |
| "grad_norm": 0.061441030353307724, |
| "learning_rate": 3.950495049504951e-05, |
| "loss": 0.0051, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.31695721077654515, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.999405562877655, |
| "eval_loss": 0.005679984577000141, |
| "eval_runtime": 32.9416, |
| "eval_samples_per_second": 153.21, |
| "eval_steps_per_second": 4.796, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.39619651347068147, |
| "grad_norm": 0.09473396837711334, |
| "learning_rate": 4.9405940594059405e-05, |
| "loss": 0.0068, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.39619651347068147, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 1.0, |
| "eval_loss": 0.008005364798009396, |
| "eval_runtime": 32.9704, |
| "eval_samples_per_second": 153.077, |
| "eval_steps_per_second": 4.792, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.4754358161648177, |
| "grad_norm": 2.07448148727417, |
| "learning_rate": 4.8965441338322695e-05, |
| "loss": 0.0078, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.4754358161648177, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9998018741607666, |
| "eval_loss": 0.007311029359698296, |
| "eval_runtime": 33.4995, |
| "eval_samples_per_second": 150.659, |
| "eval_steps_per_second": 4.716, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.554675118858954, |
| "grad_norm": 0.44440600275993347, |
| "learning_rate": 4.7864847017389395e-05, |
| "loss": 0.0066, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.554675118858954, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.999405562877655, |
| "eval_loss": 0.00900042150169611, |
| "eval_runtime": 33.8286, |
| "eval_samples_per_second": 149.193, |
| "eval_steps_per_second": 4.671, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.6339144215530903, |
| "grad_norm": 0.06913918256759644, |
| "learning_rate": 4.676425269645609e-05, |
| "loss": 0.0081, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.6339144215530903, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9996037483215332, |
| "eval_loss": 0.0071762725710868835, |
| "eval_runtime": 33.3998, |
| "eval_samples_per_second": 151.109, |
| "eval_steps_per_second": 4.731, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.7131537242472267, |
| "grad_norm": 0.7136378884315491, |
| "learning_rate": 4.566365837552278e-05, |
| "loss": 0.0096, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.7131537242472267, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.999405562877655, |
| "eval_loss": 0.008276536129415035, |
| "eval_runtime": 33.755, |
| "eval_samples_per_second": 149.519, |
| "eval_steps_per_second": 4.681, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.7923930269413629, |
| "grad_norm": 0.5450271964073181, |
| "learning_rate": 4.456306405458948e-05, |
| "loss": 0.0088, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.7923930269413629, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.999405562877655, |
| "eval_loss": 0.006255414802581072, |
| "eval_runtime": 33.084, |
| "eval_samples_per_second": 152.551, |
| "eval_steps_per_second": 4.776, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.8716323296354992, |
| "grad_norm": 0.026704631745815277, |
| "learning_rate": 4.346246973365617e-05, |
| "loss": 0.0087, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.8716323296354992, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.999405562877655, |
| "eval_loss": 0.006418165750801563, |
| "eval_runtime": 32.6589, |
| "eval_samples_per_second": 154.537, |
| "eval_steps_per_second": 4.838, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.9508716323296355, |
| "grad_norm": 2.8012053966522217, |
| "learning_rate": 4.236187541272287e-05, |
| "loss": 0.0096, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.9508716323296355, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9996037483215332, |
| "eval_loss": 0.006508174352347851, |
| "eval_runtime": 33.0205, |
| "eval_samples_per_second": 152.844, |
| "eval_steps_per_second": 4.785, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.0301109350237718, |
| "grad_norm": 0.059912703931331635, |
| "learning_rate": 4.1261281091789564e-05, |
| "loss": 0.0085, |
| "step": 1300 |
| }, |
| { |
| "epoch": 1.0301109350237718, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9998018741607666, |
| "eval_loss": 0.005652877036482096, |
| "eval_runtime": 32.9331, |
| "eval_samples_per_second": 153.25, |
| "eval_steps_per_second": 4.798, |
| "step": 1300 |
| }, |
| { |
| "epoch": 1.109350237717908, |
| "grad_norm": 0.039631109684705734, |
| "learning_rate": 4.0160686770856264e-05, |
| "loss": 0.0049, |
| "step": 1400 |
| }, |
| { |
| "epoch": 1.109350237717908, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9998018741607666, |
| "eval_loss": 0.004980658181011677, |
| "eval_runtime": 34.2128, |
| "eval_samples_per_second": 147.518, |
| "eval_steps_per_second": 4.618, |
| "step": 1400 |
| }, |
| { |
| "epoch": 1.1885895404120443, |
| "grad_norm": 0.02145099826157093, |
| "learning_rate": 3.9060092449922957e-05, |
| "loss": 0.0048, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.1885895404120443, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9996037483215332, |
| "eval_loss": 0.005376764573156834, |
| "eval_runtime": 32.9215, |
| "eval_samples_per_second": 153.304, |
| "eval_steps_per_second": 4.799, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.2678288431061806, |
| "grad_norm": 1.3420361280441284, |
| "learning_rate": 3.7959498128989656e-05, |
| "loss": 0.0032, |
| "step": 1600 |
| }, |
| { |
| "epoch": 1.2678288431061806, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9998018741607666, |
| "eval_loss": 0.005436555948108435, |
| "eval_runtime": 66.7901, |
| "eval_samples_per_second": 75.565, |
| "eval_steps_per_second": 2.366, |
| "step": 1600 |
| }, |
| { |
| "epoch": 1.3470681458003169, |
| "grad_norm": 0.040875934064388275, |
| "learning_rate": 3.685890380805635e-05, |
| "loss": 0.0017, |
| "step": 1700 |
| }, |
| { |
| "epoch": 1.3470681458003169, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.999405562877655, |
| "eval_loss": 0.006017347332090139, |
| "eval_runtime": 67.8654, |
| "eval_samples_per_second": 74.368, |
| "eval_steps_per_second": 2.328, |
| "step": 1700 |
| }, |
| { |
| "epoch": 1.4263074484944531, |
| "grad_norm": 0.030839553102850914, |
| "learning_rate": 3.575830948712305e-05, |
| "loss": 0.0032, |
| "step": 1800 |
| }, |
| { |
| "epoch": 1.4263074484944531, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.999405562877655, |
| "eval_loss": 0.0059111895971000195, |
| "eval_runtime": 67.895, |
| "eval_samples_per_second": 74.335, |
| "eval_steps_per_second": 2.327, |
| "step": 1800 |
| }, |
| { |
| "epoch": 1.5055467511885894, |
| "grad_norm": 0.06008416414260864, |
| "learning_rate": 3.465771516618974e-05, |
| "loss": 0.0072, |
| "step": 1900 |
| }, |
| { |
| "epoch": 1.5055467511885894, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9996037483215332, |
| "eval_loss": 0.006080140359699726, |
| "eval_runtime": 67.931, |
| "eval_samples_per_second": 74.296, |
| "eval_steps_per_second": 2.326, |
| "step": 1900 |
| }, |
| { |
| "epoch": 1.5847860538827259, |
| "grad_norm": 0.02880307100713253, |
| "learning_rate": 3.355712084525644e-05, |
| "loss": 0.0077, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.5847860538827259, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.999405562877655, |
| "eval_loss": 0.007432046812027693, |
| "eval_runtime": 67.8369, |
| "eval_samples_per_second": 74.399, |
| "eval_steps_per_second": 2.329, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.6640253565768621, |
| "grad_norm": 0.1070467010140419, |
| "learning_rate": 3.245652652432314e-05, |
| "loss": 0.0068, |
| "step": 2100 |
| }, |
| { |
| "epoch": 1.6640253565768621, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9952446818351746, |
| "eval_loss": 0.08792955428361893, |
| "eval_runtime": 67.8539, |
| "eval_samples_per_second": 74.38, |
| "eval_steps_per_second": 2.329, |
| "step": 2100 |
| }, |
| { |
| "epoch": 1.7432646592709984, |
| "grad_norm": 0.05642708018422127, |
| "learning_rate": 3.135593220338983e-05, |
| "loss": 0.0056, |
| "step": 2200 |
| }, |
| { |
| "epoch": 1.7432646592709984, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9996037483215332, |
| "eval_loss": 0.006125820800662041, |
| "eval_runtime": 67.8381, |
| "eval_samples_per_second": 74.398, |
| "eval_steps_per_second": 2.329, |
| "step": 2200 |
| }, |
| { |
| "epoch": 1.8225039619651349, |
| "grad_norm": 0.04333237186074257, |
| "learning_rate": 3.0255337882456532e-05, |
| "loss": 0.0087, |
| "step": 2300 |
| }, |
| { |
| "epoch": 1.8225039619651349, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 1.0, |
| "eval_loss": 0.005210440140217543, |
| "eval_runtime": 68.0258, |
| "eval_samples_per_second": 74.192, |
| "eval_steps_per_second": 2.323, |
| "step": 2300 |
| }, |
| { |
| "epoch": 1.9017432646592711, |
| "grad_norm": 0.10529103130102158, |
| "learning_rate": 2.9154743561523225e-05, |
| "loss": 0.0112, |
| "step": 2400 |
| }, |
| { |
| "epoch": 1.9017432646592711, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9998018741607666, |
| "eval_loss": 0.004964636173099279, |
| "eval_runtime": 67.9671, |
| "eval_samples_per_second": 74.257, |
| "eval_steps_per_second": 2.325, |
| "step": 2400 |
| }, |
| { |
| "epoch": 1.9809825673534074, |
| "grad_norm": 0.030080392956733704, |
| "learning_rate": 2.805414924058992e-05, |
| "loss": 0.0036, |
| "step": 2500 |
| }, |
| { |
| "epoch": 1.9809825673534074, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.999405562877655, |
| "eval_loss": 0.003934075124561787, |
| "eval_runtime": 67.9768, |
| "eval_samples_per_second": 74.246, |
| "eval_steps_per_second": 2.324, |
| "step": 2500 |
| }, |
| { |
| "epoch": 2.0602218700475436, |
| "grad_norm": 0.05913154035806656, |
| "learning_rate": 2.6953554919656613e-05, |
| "loss": 0.0047, |
| "step": 2600 |
| }, |
| { |
| "epoch": 2.0602218700475436, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.999405562877655, |
| "eval_loss": 0.004718616604804993, |
| "eval_runtime": 68.1157, |
| "eval_samples_per_second": 74.095, |
| "eval_steps_per_second": 2.32, |
| "step": 2600 |
| }, |
| { |
| "epoch": 2.13946117274168, |
| "grad_norm": 0.02771185152232647, |
| "learning_rate": 2.5852960598723313e-05, |
| "loss": 0.0054, |
| "step": 2700 |
| }, |
| { |
| "epoch": 2.13946117274168, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9998018741607666, |
| "eval_loss": 0.007168024778366089, |
| "eval_runtime": 67.8508, |
| "eval_samples_per_second": 74.384, |
| "eval_steps_per_second": 2.329, |
| "step": 2700 |
| }, |
| { |
| "epoch": 2.218700475435816, |
| "grad_norm": 0.023437298834323883, |
| "learning_rate": 2.4752366277790006e-05, |
| "loss": 0.0052, |
| "step": 2800 |
| }, |
| { |
| "epoch": 2.218700475435816, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9998018741607666, |
| "eval_loss": 0.004668584559112787, |
| "eval_runtime": 67.8782, |
| "eval_samples_per_second": 74.354, |
| "eval_steps_per_second": 2.328, |
| "step": 2800 |
| }, |
| { |
| "epoch": 2.2979397781299524, |
| "grad_norm": 0.11916761100292206, |
| "learning_rate": 2.36517719568567e-05, |
| "loss": 0.0044, |
| "step": 2900 |
| }, |
| { |
| "epoch": 2.2979397781299524, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9996037483215332, |
| "eval_loss": 0.0058599598705768585, |
| "eval_runtime": 67.9158, |
| "eval_samples_per_second": 74.313, |
| "eval_steps_per_second": 2.326, |
| "step": 2900 |
| }, |
| { |
| "epoch": 2.3771790808240887, |
| "grad_norm": 0.10791371762752533, |
| "learning_rate": 2.2551177635923398e-05, |
| "loss": 0.0051, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.3771790808240887, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9996037483215332, |
| "eval_loss": 0.00463093351572752, |
| "eval_runtime": 68.0279, |
| "eval_samples_per_second": 74.19, |
| "eval_steps_per_second": 2.323, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.456418383518225, |
| "grad_norm": 6.349626064300537, |
| "learning_rate": 2.1450583314990097e-05, |
| "loss": 0.0068, |
| "step": 3100 |
| }, |
| { |
| "epoch": 2.456418383518225, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9996037483215332, |
| "eval_loss": 0.008199479430913925, |
| "eval_runtime": 68.0218, |
| "eval_samples_per_second": 74.197, |
| "eval_steps_per_second": 2.323, |
| "step": 3100 |
| }, |
| { |
| "epoch": 2.535657686212361, |
| "grad_norm": 0.07518544048070908, |
| "learning_rate": 2.0349988994056793e-05, |
| "loss": 0.0051, |
| "step": 3200 |
| }, |
| { |
| "epoch": 2.535657686212361, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9996037483215332, |
| "eval_loss": 0.004618997685611248, |
| "eval_runtime": 67.9844, |
| "eval_samples_per_second": 74.238, |
| "eval_steps_per_second": 2.324, |
| "step": 3200 |
| }, |
| { |
| "epoch": 2.6148969889064975, |
| "grad_norm": 0.6720037460327148, |
| "learning_rate": 1.924939467312349e-05, |
| "loss": 0.0025, |
| "step": 3300 |
| }, |
| { |
| "epoch": 2.6148969889064975, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9998018741607666, |
| "eval_loss": 0.005006096325814724, |
| "eval_runtime": 67.9382, |
| "eval_samples_per_second": 74.288, |
| "eval_steps_per_second": 2.326, |
| "step": 3300 |
| }, |
| { |
| "epoch": 2.6941362916006337, |
| "grad_norm": 0.0643150731921196, |
| "learning_rate": 1.8148800352190185e-05, |
| "loss": 0.004, |
| "step": 3400 |
| }, |
| { |
| "epoch": 2.6941362916006337, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9992074370384216, |
| "eval_loss": 0.005211703013628721, |
| "eval_runtime": 67.9947, |
| "eval_samples_per_second": 74.226, |
| "eval_steps_per_second": 2.324, |
| "step": 3400 |
| }, |
| { |
| "epoch": 2.7733755942947704, |
| "grad_norm": 0.07725568860769272, |
| "learning_rate": 1.704820603125688e-05, |
| "loss": 0.0019, |
| "step": 3500 |
| }, |
| { |
| "epoch": 2.7733755942947704, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9996037483215332, |
| "eval_loss": 0.004846959374845028, |
| "eval_runtime": 68.0179, |
| "eval_samples_per_second": 74.201, |
| "eval_steps_per_second": 2.323, |
| "step": 3500 |
| }, |
| { |
| "epoch": 2.8526148969889062, |
| "grad_norm": 0.024744508787989616, |
| "learning_rate": 1.5947611710323578e-05, |
| "loss": 0.0039, |
| "step": 3600 |
| }, |
| { |
| "epoch": 2.8526148969889062, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 1.0, |
| "eval_loss": 0.0042335595935583115, |
| "eval_runtime": 68.2641, |
| "eval_samples_per_second": 73.933, |
| "eval_steps_per_second": 2.315, |
| "step": 3600 |
| }, |
| { |
| "epoch": 2.931854199683043, |
| "grad_norm": 0.015782877802848816, |
| "learning_rate": 1.4847017389390272e-05, |
| "loss": 0.0045, |
| "step": 3700 |
| }, |
| { |
| "epoch": 2.931854199683043, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9998018741607666, |
| "eval_loss": 0.004929765127599239, |
| "eval_runtime": 67.901, |
| "eval_samples_per_second": 74.329, |
| "eval_steps_per_second": 2.327, |
| "step": 3700 |
| }, |
| { |
| "epoch": 3.011093502377179, |
| "grad_norm": 2.817779779434204, |
| "learning_rate": 1.3746423068456968e-05, |
| "loss": 0.002, |
| "step": 3800 |
| }, |
| { |
| "epoch": 3.011093502377179, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9996037483215332, |
| "eval_loss": 0.004618987441062927, |
| "eval_runtime": 67.8421, |
| "eval_samples_per_second": 74.393, |
| "eval_steps_per_second": 2.329, |
| "step": 3800 |
| }, |
| { |
| "epoch": 3.0903328050713155, |
| "grad_norm": 0.011266672052443027, |
| "learning_rate": 1.2645828747523664e-05, |
| "loss": 0.0028, |
| "step": 3900 |
| }, |
| { |
| "epoch": 3.0903328050713155, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9996037483215332, |
| "eval_loss": 0.004977984819561243, |
| "eval_runtime": 67.9701, |
| "eval_samples_per_second": 74.253, |
| "eval_steps_per_second": 2.325, |
| "step": 3900 |
| }, |
| { |
| "epoch": 3.1695721077654517, |
| "grad_norm": 0.030996697023510933, |
| "learning_rate": 1.154523442659036e-05, |
| "loss": 0.0033, |
| "step": 4000 |
| }, |
| { |
| "epoch": 3.1695721077654517, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9992074370384216, |
| "eval_loss": 0.004901626612991095, |
| "eval_runtime": 68.0995, |
| "eval_samples_per_second": 74.112, |
| "eval_steps_per_second": 2.32, |
| "step": 4000 |
| }, |
| { |
| "epoch": 3.248811410459588, |
| "grad_norm": 0.011963835917413235, |
| "learning_rate": 1.0444640105657054e-05, |
| "loss": 0.0052, |
| "step": 4100 |
| }, |
| { |
| "epoch": 3.248811410459588, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9996037483215332, |
| "eval_loss": 0.0047536250203847885, |
| "eval_runtime": 67.9001, |
| "eval_samples_per_second": 74.33, |
| "eval_steps_per_second": 2.327, |
| "step": 4100 |
| }, |
| { |
| "epoch": 3.3280507131537242, |
| "grad_norm": 0.00867912545800209, |
| "learning_rate": 9.34404578472375e-06, |
| "loss": 0.0026, |
| "step": 4200 |
| }, |
| { |
| "epoch": 3.3280507131537242, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.999405562877655, |
| "eval_loss": 0.004853234626352787, |
| "eval_runtime": 67.9652, |
| "eval_samples_per_second": 74.259, |
| "eval_steps_per_second": 2.325, |
| "step": 4200 |
| }, |
| { |
| "epoch": 3.4072900158478605, |
| "grad_norm": 0.012174161151051521, |
| "learning_rate": 8.243451463790447e-06, |
| "loss": 0.0043, |
| "step": 4300 |
| }, |
| { |
| "epoch": 3.4072900158478605, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 1.0, |
| "eval_loss": 0.004362211097031832, |
| "eval_runtime": 68.0327, |
| "eval_samples_per_second": 74.185, |
| "eval_steps_per_second": 2.322, |
| "step": 4300 |
| }, |
| { |
| "epoch": 3.4865293185419968, |
| "grad_norm": 1.0564968585968018, |
| "learning_rate": 7.142857142857143e-06, |
| "loss": 0.0038, |
| "step": 4400 |
| }, |
| { |
| "epoch": 3.4865293185419968, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9998018741607666, |
| "eval_loss": 0.004115838557481766, |
| "eval_runtime": 67.8268, |
| "eval_samples_per_second": 74.41, |
| "eval_steps_per_second": 2.329, |
| "step": 4400 |
| }, |
| { |
| "epoch": 3.565768621236133, |
| "grad_norm": 0.018350793048739433, |
| "learning_rate": 6.04226282192384e-06, |
| "loss": 0.003, |
| "step": 4500 |
| }, |
| { |
| "epoch": 3.565768621236133, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 1.0, |
| "eval_loss": 0.0043433657847344875, |
| "eval_runtime": 68.0255, |
| "eval_samples_per_second": 74.193, |
| "eval_steps_per_second": 2.323, |
| "step": 4500 |
| }, |
| { |
| "epoch": 3.6450079239302693, |
| "grad_norm": 0.0198595579713583, |
| "learning_rate": 4.941668500990535e-06, |
| "loss": 0.003, |
| "step": 4600 |
| }, |
| { |
| "epoch": 3.6450079239302693, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 1.0, |
| "eval_loss": 0.004538228269666433, |
| "eval_runtime": 67.8043, |
| "eval_samples_per_second": 74.435, |
| "eval_steps_per_second": 2.33, |
| "step": 4600 |
| }, |
| { |
| "epoch": 3.7242472266244055, |
| "grad_norm": 0.014352944679558277, |
| "learning_rate": 3.841074180057231e-06, |
| "loss": 0.003, |
| "step": 4700 |
| }, |
| { |
| "epoch": 3.7242472266244055, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9998018741607666, |
| "eval_loss": 0.004474899731576443, |
| "eval_runtime": 67.8863, |
| "eval_samples_per_second": 74.345, |
| "eval_steps_per_second": 2.327, |
| "step": 4700 |
| }, |
| { |
| "epoch": 3.8034865293185423, |
| "grad_norm": 0.05333567038178444, |
| "learning_rate": 2.740479859123927e-06, |
| "loss": 0.0009, |
| "step": 4800 |
| }, |
| { |
| "epoch": 3.8034865293185423, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9998018741607666, |
| "eval_loss": 0.0041495212353765965, |
| "eval_runtime": 68.1146, |
| "eval_samples_per_second": 74.096, |
| "eval_steps_per_second": 2.32, |
| "step": 4800 |
| }, |
| { |
| "epoch": 3.882725832012678, |
| "grad_norm": 0.017361685633659363, |
| "learning_rate": 1.639885538190623e-06, |
| "loss": 0.0048, |
| "step": 4900 |
| }, |
| { |
| "epoch": 3.882725832012678, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9998018741607666, |
| "eval_loss": 0.004217915236949921, |
| "eval_runtime": 67.5299, |
| "eval_samples_per_second": 74.737, |
| "eval_steps_per_second": 2.34, |
| "step": 4900 |
| }, |
| { |
| "epoch": 3.9619651347068148, |
| "grad_norm": 0.021525979042053223, |
| "learning_rate": 5.392912172573191e-07, |
| "loss": 0.0035, |
| "step": 5000 |
| }, |
| { |
| "epoch": 3.9619651347068148, |
| "eval_Embedding_Dataset_Dev_cosine_accuracy": 0.9996037483215332, |
| "eval_loss": 0.004212545696645975, |
| "eval_runtime": 67.678, |
| "eval_samples_per_second": 74.574, |
| "eval_steps_per_second": 2.335, |
| "step": 5000 |
| } |
| ], |
| "logging_steps": 100, |
| "max_steps": 5048, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 4, |
| "save_steps": 100, |
| "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": 32, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
