| { | |
| "best_metric": 0.637806236743927, | |
| "best_model_checkpoint": "/data/jcanete/all_results/xnli/albeto_tiny2/epochs_3_bs_16_lr_3e-5/checkpoint-44000", | |
| "epoch": 3.0, | |
| "global_step": 73632, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 2.9797506518904825e-05, | |
| "loss": 0.9759, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 2.959379074315515e-05, | |
| "loss": 0.8759, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 2.9390074967405478e-05, | |
| "loss": 0.8565, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 2.9186359191655804e-05, | |
| "loss": 0.8281, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_accuracy": 0.6373493975903615, | |
| "eval_loss": 0.8259181976318359, | |
| "eval_runtime": 0.5925, | |
| "eval_samples_per_second": 4202.704, | |
| "eval_steps_per_second": 263.302, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 2.8983050847457626e-05, | |
| "loss": 0.8181, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 2.8779335071707956e-05, | |
| "loss": 0.8103, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 2.857561929595828e-05, | |
| "loss": 0.7937, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 2.8371903520208605e-05, | |
| "loss": 0.772, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_accuracy": 0.6738955823293172, | |
| "eval_loss": 0.7611338496208191, | |
| "eval_runtime": 0.5925, | |
| "eval_samples_per_second": 4202.255, | |
| "eval_steps_per_second": 263.274, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 2.816818774445893e-05, | |
| "loss": 0.7804, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 2.7964471968709258e-05, | |
| "loss": 0.7733, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 2.7760756192959584e-05, | |
| "loss": 0.7669, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 2.755744784876141e-05, | |
| "loss": 0.7516, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_accuracy": 0.6759036144578313, | |
| "eval_loss": 0.7518651485443115, | |
| "eval_runtime": 0.5928, | |
| "eval_samples_per_second": 4200.064, | |
| "eval_steps_per_second": 263.137, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 2.7353732073011732e-05, | |
| "loss": 0.7505, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 2.7150016297262062e-05, | |
| "loss": 0.7512, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 2.6946300521512385e-05, | |
| "loss": 0.7414, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 2.6742584745762715e-05, | |
| "loss": 0.7311, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_accuracy": 0.672289156626506, | |
| "eval_loss": 0.7527135610580444, | |
| "eval_runtime": 0.5934, | |
| "eval_samples_per_second": 4195.935, | |
| "eval_steps_per_second": 262.878, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 2.6538868970013038e-05, | |
| "loss": 0.723, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 2.6335560625814863e-05, | |
| "loss": 0.7138, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 2.613184485006519e-05, | |
| "loss": 0.7198, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 2.5928129074315516e-05, | |
| "loss": 0.7066, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "eval_accuracy": 0.6975903614457831, | |
| "eval_loss": 0.7133071422576904, | |
| "eval_runtime": 0.5925, | |
| "eval_samples_per_second": 4202.512, | |
| "eval_steps_per_second": 263.29, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 2.5724413298565842e-05, | |
| "loss": 0.7101, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 2.552069752281617e-05, | |
| "loss": 0.7095, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 2.5317389178617994e-05, | |
| "loss": 0.7133, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 2.511367340286832e-05, | |
| "loss": 0.7008, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_accuracy": 0.7092369477911646, | |
| "eval_loss": 0.7024465203285217, | |
| "eval_runtime": 0.5924, | |
| "eval_samples_per_second": 4203.393, | |
| "eval_steps_per_second": 263.345, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 2.4909957627118643e-05, | |
| "loss": 0.7074, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 2.470624185136897e-05, | |
| "loss": 0.7082, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 2.4502526075619296e-05, | |
| "loss": 0.7128, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 2.429921773142112e-05, | |
| "loss": 0.704, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_accuracy": 0.7040160642570281, | |
| "eval_loss": 0.7015809416770935, | |
| "eval_runtime": 0.5907, | |
| "eval_samples_per_second": 4215.675, | |
| "eval_steps_per_second": 264.115, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 2.4095501955671448e-05, | |
| "loss": 0.6859, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 2.3891786179921774e-05, | |
| "loss": 0.7053, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 2.36880704041721e-05, | |
| "loss": 0.6885, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 2.3484762059973926e-05, | |
| "loss": 0.6942, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_accuracy": 0.7088353413654619, | |
| "eval_loss": 0.6795119047164917, | |
| "eval_runtime": 0.5915, | |
| "eval_samples_per_second": 4209.383, | |
| "eval_steps_per_second": 263.72, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 2.3281046284224252e-05, | |
| "loss": 0.6902, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 2.3077330508474575e-05, | |
| "loss": 0.6819, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 2.2873614732724905e-05, | |
| "loss": 0.6905, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 2.2669898956975228e-05, | |
| "loss": 0.6714, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_accuracy": 0.6991967871485943, | |
| "eval_loss": 0.7053298354148865, | |
| "eval_runtime": 0.5927, | |
| "eval_samples_per_second": 4201.418, | |
| "eval_steps_per_second": 263.221, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 2.2466183181225557e-05, | |
| "loss": 0.6783, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 2.226246740547588e-05, | |
| "loss": 0.6871, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 2.2059159061277706e-05, | |
| "loss": 0.6805, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 2.1855443285528032e-05, | |
| "loss": 0.6607, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "eval_accuracy": 0.7152610441767068, | |
| "eval_loss": 0.6862269639968872, | |
| "eval_runtime": 0.5938, | |
| "eval_samples_per_second": 4193.683, | |
| "eval_steps_per_second": 262.737, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 2.165172750977836e-05, | |
| "loss": 0.6743, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 2.1448011734028685e-05, | |
| "loss": 0.6743, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 2.124470338983051e-05, | |
| "loss": 0.6648, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 2.1040987614080833e-05, | |
| "loss": 0.6815, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_accuracy": 0.7253012048192771, | |
| "eval_loss": 0.6604429483413696, | |
| "eval_runtime": 0.5927, | |
| "eval_samples_per_second": 4200.88, | |
| "eval_steps_per_second": 263.188, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 2.0837271838331163e-05, | |
| "loss": 0.6666, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 2.0633556062581486e-05, | |
| "loss": 0.6658, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 2.0429840286831812e-05, | |
| "loss": 0.666, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 2.022612451108214e-05, | |
| "loss": 0.6773, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_accuracy": 0.7253012048192771, | |
| "eval_loss": 0.6634726524353027, | |
| "eval_runtime": 0.5908, | |
| "eval_samples_per_second": 4214.626, | |
| "eval_steps_per_second": 264.049, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 2.0022816166883964e-05, | |
| "loss": 0.6609, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 1.9819100391134287e-05, | |
| "loss": 0.6014, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 1.9615384615384617e-05, | |
| "loss": 0.6089, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 1.941166883963494e-05, | |
| "loss": 0.6011, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "eval_accuracy": 0.7281124497991968, | |
| "eval_loss": 0.6600359678268433, | |
| "eval_runtime": 0.5949, | |
| "eval_samples_per_second": 4185.679, | |
| "eval_steps_per_second": 262.235, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 1.920795306388527e-05, | |
| "loss": 0.6012, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 1.9004644719687095e-05, | |
| "loss": 0.5788, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 1.8800928943937418e-05, | |
| "loss": 0.6058, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 1.8597213168187744e-05, | |
| "loss": 0.6154, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_accuracy": 0.7309236947791165, | |
| "eval_loss": 0.6442546844482422, | |
| "eval_runtime": 0.5944, | |
| "eval_samples_per_second": 4189.183, | |
| "eval_steps_per_second": 262.455, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 1.839349739243807e-05, | |
| "loss": 0.6147, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 1.8189781616688397e-05, | |
| "loss": 0.5903, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 1.7986473272490222e-05, | |
| "loss": 0.6156, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 1.778275749674055e-05, | |
| "loss": 0.6024, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "eval_accuracy": 0.7188755020080321, | |
| "eval_loss": 0.6576093435287476, | |
| "eval_runtime": 0.5967, | |
| "eval_samples_per_second": 4172.848, | |
| "eval_steps_per_second": 261.431, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 1.7579041720990875e-05, | |
| "loss": 0.5907, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 1.73753259452412e-05, | |
| "loss": 0.6012, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 1.7172017601043023e-05, | |
| "loss": 0.5961, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1.6968301825293353e-05, | |
| "loss": 0.5971, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_accuracy": 0.7325301204819277, | |
| "eval_loss": 0.6447591185569763, | |
| "eval_runtime": 0.5962, | |
| "eval_samples_per_second": 4176.317, | |
| "eval_steps_per_second": 261.649, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 1.6764586049543676e-05, | |
| "loss": 0.5961, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 1.6560870273794005e-05, | |
| "loss": 0.5898, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 1.6357561929595828e-05, | |
| "loss": 0.5923, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 1.6153846153846154e-05, | |
| "loss": 0.6002, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "eval_accuracy": 0.7333333333333333, | |
| "eval_loss": 0.6481831669807434, | |
| "eval_runtime": 0.5959, | |
| "eval_samples_per_second": 4178.807, | |
| "eval_steps_per_second": 261.805, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 1.595013037809648e-05, | |
| "loss": 0.5887, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 1.5746414602346807e-05, | |
| "loss": 0.595, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 1.5543106258148632e-05, | |
| "loss": 0.5868, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 1.533939048239896e-05, | |
| "loss": 0.5813, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "eval_accuracy": 0.7265060240963855, | |
| "eval_loss": 0.654254138469696, | |
| "eval_runtime": 0.5952, | |
| "eval_samples_per_second": 4183.66, | |
| "eval_steps_per_second": 262.109, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 1.5135674706649283e-05, | |
| "loss": 0.5891, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 1.493195893089961e-05, | |
| "loss": 0.5905, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 1.4728650586701433e-05, | |
| "loss": 0.5853, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 1.452493481095176e-05, | |
| "loss": 0.5962, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "eval_accuracy": 0.738152610441767, | |
| "eval_loss": 0.6403775215148926, | |
| "eval_runtime": 0.5945, | |
| "eval_samples_per_second": 4188.254, | |
| "eval_steps_per_second": 262.397, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.4321219035202086e-05, | |
| "loss": 0.609, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 1.4117503259452412e-05, | |
| "loss": 0.5753, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 1.3914194915254236e-05, | |
| "loss": 0.5903, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 1.3710479139504562e-05, | |
| "loss": 0.5948, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "eval_accuracy": 0.7224899598393574, | |
| "eval_loss": 0.6676555871963501, | |
| "eval_runtime": 0.597, | |
| "eval_samples_per_second": 4170.973, | |
| "eval_steps_per_second": 261.314, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 1.3506763363754889e-05, | |
| "loss": 0.5847, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 1.3303047588005215e-05, | |
| "loss": 0.5869, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 1.309973924380704e-05, | |
| "loss": 0.5945, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 1.2896023468057367e-05, | |
| "loss": 0.5975, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "eval_accuracy": 0.7253012048192771, | |
| "eval_loss": 0.6549131870269775, | |
| "eval_runtime": 0.5935, | |
| "eval_samples_per_second": 4195.227, | |
| "eval_steps_per_second": 262.834, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 1.2692307692307693e-05, | |
| "loss": 0.5846, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 1.248859191655802e-05, | |
| "loss": 0.5907, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 1.2284876140808344e-05, | |
| "loss": 0.5929, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 1.208116036505867e-05, | |
| "loss": 0.5911, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "eval_accuracy": 0.738152610441767, | |
| "eval_loss": 0.637806236743927, | |
| "eval_runtime": 0.5938, | |
| "eval_samples_per_second": 4193.465, | |
| "eval_steps_per_second": 262.723, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 1.1877444589308995e-05, | |
| "loss": 0.5842, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 1.1673728813559321e-05, | |
| "loss": 0.5929, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 1.1470420469361149e-05, | |
| "loss": 0.5767, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 1.1266704693611475e-05, | |
| "loss": 0.5763, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_accuracy": 0.7293172690763052, | |
| "eval_loss": 0.640717089176178, | |
| "eval_runtime": 0.5976, | |
| "eval_samples_per_second": 4166.642, | |
| "eval_steps_per_second": 261.043, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 1.10629889178618e-05, | |
| "loss": 0.5852, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 1.0859273142112126e-05, | |
| "loss": 0.5845, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 1.0655964797913951e-05, | |
| "loss": 0.5817, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 1.0452249022164278e-05, | |
| "loss": 0.5767, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "eval_accuracy": 0.7220883534136546, | |
| "eval_loss": 0.653072714805603, | |
| "eval_runtime": 0.588, | |
| "eval_samples_per_second": 4234.632, | |
| "eval_steps_per_second": 265.302, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 1.0248533246414602e-05, | |
| "loss": 0.5832, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.0044817470664928e-05, | |
| "loss": 0.5842, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 9.841509126466754e-06, | |
| "loss": 0.5185, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 9.637793350717079e-06, | |
| "loss": 0.5249, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_accuracy": 0.7317269076305221, | |
| "eval_loss": 0.6527847647666931, | |
| "eval_runtime": 0.5908, | |
| "eval_samples_per_second": 4214.33, | |
| "eval_steps_per_second": 264.03, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 9.434077574967405e-06, | |
| "loss": 0.5192, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 9.230361799217731e-06, | |
| "loss": 0.5124, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 9.026646023468058e-06, | |
| "loss": 0.5136, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 8.823337679269881e-06, | |
| "loss": 0.5156, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "eval_accuracy": 0.7309236947791165, | |
| "eval_loss": 0.6581133604049683, | |
| "eval_runtime": 0.5908, | |
| "eval_samples_per_second": 4214.598, | |
| "eval_steps_per_second": 264.047, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 8.619621903520208e-06, | |
| "loss": 0.5059, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 8.415906127770534e-06, | |
| "loss": 0.496, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 8.21219035202086e-06, | |
| "loss": 0.5136, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 8.008882007822686e-06, | |
| "loss": 0.507, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "eval_accuracy": 0.734136546184739, | |
| "eval_loss": 0.6511837244033813, | |
| "eval_runtime": 0.5907, | |
| "eval_samples_per_second": 4215.648, | |
| "eval_steps_per_second": 264.113, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 7.805166232073012e-06, | |
| "loss": 0.5093, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 7.601450456323338e-06, | |
| "loss": 0.5042, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 7.397734680573664e-06, | |
| "loss": 0.5008, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 7.19401890482399e-06, | |
| "loss": 0.5033, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "eval_accuracy": 0.7385542168674699, | |
| "eval_loss": 0.6447919607162476, | |
| "eval_runtime": 0.5923, | |
| "eval_samples_per_second": 4203.946, | |
| "eval_steps_per_second": 263.38, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 6.990710560625815e-06, | |
| "loss": 0.511, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 6.786994784876141e-06, | |
| "loss": 0.5091, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 6.583279009126467e-06, | |
| "loss": 0.5207, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 6.379563233376792e-06, | |
| "loss": 0.5174, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "eval_accuracy": 0.7313253012048193, | |
| "eval_loss": 0.6609730124473572, | |
| "eval_runtime": 0.5899, | |
| "eval_samples_per_second": 4221.14, | |
| "eval_steps_per_second": 264.457, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 6.176254889178618e-06, | |
| "loss": 0.5105, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 5.972539113428944e-06, | |
| "loss": 0.5024, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 5.7688233376792704e-06, | |
| "loss": 0.5061, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 5.565107561929596e-06, | |
| "loss": 0.5, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "eval_accuracy": 0.7333333333333333, | |
| "eval_loss": 0.6551468372344971, | |
| "eval_runtime": 0.59, | |
| "eval_samples_per_second": 4220.341, | |
| "eval_steps_per_second": 264.407, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 5.3617992177314215e-06, | |
| "loss": 0.5108, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 5.158083441981748e-06, | |
| "loss": 0.5017, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 4.954367666232072e-06, | |
| "loss": 0.5192, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 4.750651890482399e-06, | |
| "loss": 0.5016, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "eval_accuracy": 0.7405622489959839, | |
| "eval_loss": 0.6423055529594421, | |
| "eval_runtime": 0.5906, | |
| "eval_samples_per_second": 4216.028, | |
| "eval_steps_per_second": 264.137, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 4.547343546284224e-06, | |
| "loss": 0.5066, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 4.343627770534551e-06, | |
| "loss": 0.5017, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 4.140319426336375e-06, | |
| "loss": 0.5159, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 3.936603650586702e-06, | |
| "loss": 0.5105, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "eval_accuracy": 0.7329317269076305, | |
| "eval_loss": 0.6546612977981567, | |
| "eval_runtime": 0.5911, | |
| "eval_samples_per_second": 4212.749, | |
| "eval_steps_per_second": 263.931, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 3.7328878748370275e-06, | |
| "loss": 0.5042, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 3.5291720990873534e-06, | |
| "loss": 0.5025, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 3.3254563233376792e-06, | |
| "loss": 0.5043, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 3.1217405475880055e-06, | |
| "loss": 0.5091, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "eval_accuracy": 0.7305220883534137, | |
| "eval_loss": 0.6591814756393433, | |
| "eval_runtime": 0.5919, | |
| "eval_samples_per_second": 4206.957, | |
| "eval_steps_per_second": 263.568, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 2.918024771838331e-06, | |
| "loss": 0.5049, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 2.7143089960886573e-06, | |
| "loss": 0.4968, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 2.5110006518904825e-06, | |
| "loss": 0.5058, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 2.3072848761408087e-06, | |
| "loss": 0.5142, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_accuracy": 0.7401606425702811, | |
| "eval_loss": 0.648106575012207, | |
| "eval_runtime": 0.5937, | |
| "eval_samples_per_second": 4194.216, | |
| "eval_steps_per_second": 262.77, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 2.103569100391134e-06, | |
| "loss": 0.4923, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 1.89985332464146e-06, | |
| "loss": 0.5092, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 1.6965449804432857e-06, | |
| "loss": 0.5197, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 1.4928292046936115e-06, | |
| "loss": 0.5088, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "eval_accuracy": 0.7389558232931727, | |
| "eval_loss": 0.6492660641670227, | |
| "eval_runtime": 0.5917, | |
| "eval_samples_per_second": 4207.954, | |
| "eval_steps_per_second": 263.631, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 1.2891134289439376e-06, | |
| "loss": 0.5246, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 1.0853976531942633e-06, | |
| "loss": 0.5053, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 8.816818774445892e-07, | |
| "loss": 0.5007, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 6.783735332464145e-07, | |
| "loss": 0.5012, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "eval_accuracy": 0.7417670682730924, | |
| "eval_loss": 0.6490946412086487, | |
| "eval_runtime": 0.591, | |
| "eval_samples_per_second": 4213.476, | |
| "eval_steps_per_second": 263.977, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 4.7465775749674057e-07, | |
| "loss": 0.5039, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 2.7094198174706654e-07, | |
| "loss": 0.5009, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 6.722620599739244e-08, | |
| "loss": 0.5134, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 73632, | |
| "total_flos": 809331840200640.0, | |
| "train_loss": 0.6086252490005924, | |
| "train_runtime": 925.6681, | |
| "train_samples_per_second": 1272.709, | |
| "train_steps_per_second": 79.545 | |
| } | |
| ], | |
| "max_steps": 73632, | |
| "num_train_epochs": 3, | |
| "total_flos": 809331840200640.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |