| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.0, | |
| "global_step": 15596, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 1e-05, | |
| "loss": 2.9595, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 1.9980000000000002e-05, | |
| "loss": 2.8349, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 2.9959999999999998e-05, | |
| "loss": 2.5062, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_accuracy": 0.24551166594028473, | |
| "eval_loss": 2.527085304260254, | |
| "eval_runtime": 249.1452, | |
| "eval_samples_per_second": 8.943, | |
| "eval_steps_per_second": 4.471, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 3.9960000000000004e-05, | |
| "loss": 2.325, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_accuracy": 0.2921903133392334, | |
| "eval_loss": 2.155109405517578, | |
| "eval_runtime": 248.6187, | |
| "eval_samples_per_second": 8.962, | |
| "eval_steps_per_second": 4.481, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 4.996e-05, | |
| "loss": 2.0841, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_accuracy": 0.37836626172065735, | |
| "eval_loss": 1.9377281665802002, | |
| "eval_runtime": 248.5683, | |
| "eval_samples_per_second": 8.963, | |
| "eval_steps_per_second": 4.482, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 5.9940000000000005e-05, | |
| "loss": 1.8918, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "eval_accuracy": 0.44120287895202637, | |
| "eval_loss": 1.7982349395751953, | |
| "eval_runtime": 248.5588, | |
| "eval_samples_per_second": 8.964, | |
| "eval_steps_per_second": 4.482, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 6.994000000000001e-05, | |
| "loss": 1.7864, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "eval_accuracy": 0.4039497375488281, | |
| "eval_loss": 1.8799262046813965, | |
| "eval_runtime": 248.6043, | |
| "eval_samples_per_second": 8.962, | |
| "eval_steps_per_second": 4.481, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 7.994000000000001e-05, | |
| "loss": 1.6033, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "eval_accuracy": 0.4923698306083679, | |
| "eval_loss": 1.6540141105651855, | |
| "eval_runtime": 248.311, | |
| "eval_samples_per_second": 8.973, | |
| "eval_steps_per_second": 4.486, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 8.994e-05, | |
| "loss": 1.4896, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "eval_accuracy": 0.4865350127220154, | |
| "eval_loss": 1.8671057224273682, | |
| "eval_runtime": 248.1758, | |
| "eval_samples_per_second": 8.978, | |
| "eval_steps_per_second": 4.489, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 9.992e-05, | |
| "loss": 1.4354, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "eval_accuracy": 0.6077199578285217, | |
| "eval_loss": 1.2887369394302368, | |
| "eval_runtime": 285.3825, | |
| "eval_samples_per_second": 7.807, | |
| "eval_steps_per_second": 3.904, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 9.531898829747074e-05, | |
| "loss": 1.2833, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "eval_accuracy": 0.5520646572113037, | |
| "eval_loss": 1.6005043983459473, | |
| "eval_runtime": 244.6199, | |
| "eval_samples_per_second": 9.108, | |
| "eval_steps_per_second": 4.554, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 9.060022650056625e-05, | |
| "loss": 1.2117, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "eval_accuracy": 0.6261220574378967, | |
| "eval_loss": 1.3362298011779785, | |
| "eval_runtime": 243.551, | |
| "eval_samples_per_second": 9.148, | |
| "eval_steps_per_second": 4.574, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 8.588146470366177e-05, | |
| "loss": 1.1316, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "eval_accuracy": 0.6337522268295288, | |
| "eval_loss": 1.2837433815002441, | |
| "eval_runtime": 243.5454, | |
| "eval_samples_per_second": 9.148, | |
| "eval_steps_per_second": 4.574, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 8.116270290675727e-05, | |
| "loss": 1.0295, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "eval_accuracy": 0.7369838356971741, | |
| "eval_loss": 0.9227874875068665, | |
| "eval_runtime": 245.7687, | |
| "eval_samples_per_second": 9.065, | |
| "eval_steps_per_second": 4.533, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 7.646281615704039e-05, | |
| "loss": 0.9247, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "eval_accuracy": 0.71499103307724, | |
| "eval_loss": 1.092219352722168, | |
| "eval_runtime": 245.4882, | |
| "eval_samples_per_second": 9.076, | |
| "eval_steps_per_second": 4.538, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 7.174405436013591e-05, | |
| "loss": 0.89, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "eval_accuracy": 0.708707332611084, | |
| "eval_loss": 1.2390897274017334, | |
| "eval_runtime": 245.0893, | |
| "eval_samples_per_second": 9.091, | |
| "eval_steps_per_second": 4.545, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 6.702529256323142e-05, | |
| "loss": 0.8384, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "eval_accuracy": 0.8002693057060242, | |
| "eval_loss": 0.8005222082138062, | |
| "eval_runtime": 245.3523, | |
| "eval_samples_per_second": 9.081, | |
| "eval_steps_per_second": 4.54, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 6.230653076632692e-05, | |
| "loss": 0.7628, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "eval_accuracy": 0.7881507873535156, | |
| "eval_loss": 0.8995262980461121, | |
| "eval_runtime": 245.5663, | |
| "eval_samples_per_second": 9.073, | |
| "eval_steps_per_second": 4.536, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 5.758776896942243e-05, | |
| "loss": 0.6186, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "eval_accuracy": 0.7675045132637024, | |
| "eval_loss": 1.215416431427002, | |
| "eval_runtime": 245.7222, | |
| "eval_samples_per_second": 9.067, | |
| "eval_steps_per_second": 4.534, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 5.287844469611174e-05, | |
| "loss": 0.6508, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "eval_accuracy": 0.7975763082504272, | |
| "eval_loss": 1.0451174974441528, | |
| "eval_runtime": 245.3759, | |
| "eval_samples_per_second": 9.08, | |
| "eval_steps_per_second": 4.54, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "learning_rate": 4.8159682899207246e-05, | |
| "loss": 0.6042, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "eval_accuracy": 0.7939856648445129, | |
| "eval_loss": 1.1320722103118896, | |
| "eval_runtime": 246.1024, | |
| "eval_samples_per_second": 9.053, | |
| "eval_steps_per_second": 4.527, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "learning_rate": 4.344092110230276e-05, | |
| "loss": 0.4819, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "eval_accuracy": 0.8177737593650818, | |
| "eval_loss": 0.9788008332252502, | |
| "eval_runtime": 244.8325, | |
| "eval_samples_per_second": 9.1, | |
| "eval_steps_per_second": 4.55, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 3.8722159305398266e-05, | |
| "loss": 0.4216, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "eval_accuracy": 0.8482944369316101, | |
| "eval_loss": 0.8459686040878296, | |
| "eval_runtime": 245.8866, | |
| "eval_samples_per_second": 9.061, | |
| "eval_steps_per_second": 4.531, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 5.39, | |
| "learning_rate": 3.401283503208758e-05, | |
| "loss": 0.3622, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.39, | |
| "eval_accuracy": 0.8357270956039429, | |
| "eval_loss": 0.9635916948318481, | |
| "eval_runtime": 244.3222, | |
| "eval_samples_per_second": 9.119, | |
| "eval_steps_per_second": 4.56, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 2.929407323518309e-05, | |
| "loss": 0.3374, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "eval_accuracy": 0.8563734292984009, | |
| "eval_loss": 0.8446475267410278, | |
| "eval_runtime": 244.1364, | |
| "eval_samples_per_second": 9.126, | |
| "eval_steps_per_second": 4.563, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 5.83, | |
| "learning_rate": 2.4575311438278596e-05, | |
| "loss": 0.3742, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 5.83, | |
| "eval_accuracy": 0.884649932384491, | |
| "eval_loss": 0.6776938438415527, | |
| "eval_runtime": 245.0189, | |
| "eval_samples_per_second": 9.093, | |
| "eval_steps_per_second": 4.547, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 1.9856549641374103e-05, | |
| "loss": 0.3117, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "eval_accuracy": 0.8675942420959473, | |
| "eval_loss": 0.8386155366897583, | |
| "eval_runtime": 246.3484, | |
| "eval_samples_per_second": 9.044, | |
| "eval_steps_per_second": 4.522, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "learning_rate": 1.513778784446961e-05, | |
| "loss": 0.2345, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "eval_accuracy": 0.8725314140319824, | |
| "eval_loss": 0.8390255570411682, | |
| "eval_runtime": 255.0557, | |
| "eval_samples_per_second": 8.735, | |
| "eval_steps_per_second": 4.368, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "learning_rate": 1.0419026047565119e-05, | |
| "loss": 0.1739, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "eval_accuracy": 0.8994613885879517, | |
| "eval_loss": 0.6298801302909851, | |
| "eval_runtime": 257.8872, | |
| "eval_samples_per_second": 8.639, | |
| "eval_steps_per_second": 4.32, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 6.73, | |
| "learning_rate": 5.700264250660626e-06, | |
| "loss": 0.1661, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 6.73, | |
| "eval_accuracy": 0.9026032090187073, | |
| "eval_loss": 0.6533035635948181, | |
| "eval_runtime": 256.0539, | |
| "eval_samples_per_second": 8.701, | |
| "eval_steps_per_second": 4.351, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "learning_rate": 9.909399773499434e-07, | |
| "loss": 0.1712, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "eval_accuracy": 0.8940753936767578, | |
| "eval_loss": 0.7215341925621033, | |
| "eval_runtime": 252.7573, | |
| "eval_samples_per_second": 8.815, | |
| "eval_steps_per_second": 4.407, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "step": 15596, | |
| "total_flos": 3.102819690448767e+18, | |
| "train_runtime": 390.3998, | |
| "train_samples_per_second": 159.795, | |
| "train_steps_per_second": 39.949 | |
| } | |
| ], | |
| "max_steps": 15596, | |
| "num_train_epochs": 7, | |
| "total_flos": 3.102819690448767e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |