| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.328794068694205, | |
| "global_step": 330000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.7478186311595304e-05, | |
| "loss": 0.0451, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 0.04004961997270584, | |
| "eval_runtime": 280.6429, | |
| "eval_samples_per_second": 78.498, | |
| "eval_steps_per_second": 19.626, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.49563726231906e-05, | |
| "loss": 0.0479, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 0.040193621069192886, | |
| "eval_runtime": 274.5073, | |
| "eval_samples_per_second": 80.253, | |
| "eval_steps_per_second": 20.065, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.24345589347859e-05, | |
| "loss": 0.0442, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 0.040153153240680695, | |
| "eval_runtime": 270.832, | |
| "eval_samples_per_second": 81.342, | |
| "eval_steps_per_second": 20.337, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3.99127452463812e-05, | |
| "loss": 0.0443, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.04070935398340225, | |
| "eval_runtime": 270.5632, | |
| "eval_samples_per_second": 81.423, | |
| "eval_steps_per_second": 20.358, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 3.73909315579765e-05, | |
| "loss": 0.044, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.0404120609164238, | |
| "eval_runtime": 270.4511, | |
| "eval_samples_per_second": 81.456, | |
| "eval_steps_per_second": 20.366, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 3.48691178695718e-05, | |
| "loss": 0.0469, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.03903116658329964, | |
| "eval_runtime": 270.9455, | |
| "eval_samples_per_second": 81.308, | |
| "eval_steps_per_second": 20.329, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 3.23473041811671e-05, | |
| "loss": 0.0464, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_loss": 0.039902038872241974, | |
| "eval_runtime": 271.2781, | |
| "eval_samples_per_second": 81.208, | |
| "eval_steps_per_second": 20.304, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 2.9825490492762397e-05, | |
| "loss": 0.0454, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.04037521034479141, | |
| "eval_runtime": 270.9003, | |
| "eval_samples_per_second": 81.321, | |
| "eval_steps_per_second": 20.332, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 2.7303676804357698e-05, | |
| "loss": 0.0446, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_loss": 0.03930870071053505, | |
| "eval_runtime": 274.5685, | |
| "eval_samples_per_second": 80.235, | |
| "eval_steps_per_second": 20.061, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 2.4781863115952993e-05, | |
| "loss": 0.0426, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.03955301269888878, | |
| "eval_runtime": 267.684, | |
| "eval_samples_per_second": 82.299, | |
| "eval_steps_per_second": 20.576, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 2.2260049427548294e-05, | |
| "loss": 0.0444, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_loss": 0.03960525989532471, | |
| "eval_runtime": 267.9101, | |
| "eval_samples_per_second": 82.229, | |
| "eval_steps_per_second": 20.559, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.9738235739143592e-05, | |
| "loss": 0.0436, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_loss": 0.039675917476415634, | |
| "eval_runtime": 267.1715, | |
| "eval_samples_per_second": 82.456, | |
| "eval_steps_per_second": 20.616, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 1.7216422050738893e-05, | |
| "loss": 0.0438, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 0.03922617807984352, | |
| "eval_runtime": 275.6861, | |
| "eval_samples_per_second": 79.91, | |
| "eval_steps_per_second": 19.979, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.4694608362334191e-05, | |
| "loss": 0.0429, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_loss": 0.03912888839840889, | |
| "eval_runtime": 268.88, | |
| "eval_samples_per_second": 81.932, | |
| "eval_steps_per_second": 20.485, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 4.5965098098552486e-05, | |
| "loss": 0.044, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "eval_loss": 0.039720822125673294, | |
| "eval_runtime": 282.371, | |
| "eval_samples_per_second": 78.018, | |
| "eval_steps_per_second": 19.506, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.546073536087154e-05, | |
| "loss": 0.0438, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 4.49563726231906e-05, | |
| "loss": 0.0438, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_loss": 0.04048037528991699, | |
| "eval_runtime": 267.0469, | |
| "eval_samples_per_second": 82.495, | |
| "eval_steps_per_second": 20.626, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 4.445200988550966e-05, | |
| "loss": 0.0445, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 4.394764714782872e-05, | |
| "loss": 0.0439, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "eval_loss": 0.04071788489818573, | |
| "eval_runtime": 265.1859, | |
| "eval_samples_per_second": 83.074, | |
| "eval_steps_per_second": 20.77, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 4.344328441014778e-05, | |
| "loss": 0.0449, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 4.2938921672466844e-05, | |
| "loss": 0.043, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "eval_loss": 0.04017642140388489, | |
| "eval_runtime": 280.2898, | |
| "eval_samples_per_second": 78.597, | |
| "eval_steps_per_second": 19.651, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 4.24345589347859e-05, | |
| "loss": 0.0448, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 4.193019619710496e-05, | |
| "loss": 0.0434, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "eval_loss": 0.03977196663618088, | |
| "eval_runtime": 273.4991, | |
| "eval_samples_per_second": 80.549, | |
| "eval_steps_per_second": 20.139, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 4.142583345942402e-05, | |
| "loss": 0.0416, | |
| "step": 170000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 4.0921470721743075e-05, | |
| "loss": 0.041, | |
| "step": 180000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_loss": 0.04102006182074547, | |
| "eval_runtime": 281.9283, | |
| "eval_samples_per_second": 78.14, | |
| "eval_steps_per_second": 19.537, | |
| "step": 180000 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 4.041710798406214e-05, | |
| "loss": 0.0433, | |
| "step": 190000 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 3.99127452463812e-05, | |
| "loss": 0.0429, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "eval_loss": 0.03989257290959358, | |
| "eval_runtime": 280.0034, | |
| "eval_samples_per_second": 78.678, | |
| "eval_steps_per_second": 19.671, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 3.940838250870026e-05, | |
| "loss": 0.0415, | |
| "step": 210000 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 3.890401977101932e-05, | |
| "loss": 0.0414, | |
| "step": 220000 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "eval_loss": 0.04061907157301903, | |
| "eval_runtime": 267.0149, | |
| "eval_samples_per_second": 82.505, | |
| "eval_steps_per_second": 20.628, | |
| "step": 220000 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 3.8399657033338377e-05, | |
| "loss": 0.0418, | |
| "step": 230000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 3.789529429565744e-05, | |
| "loss": 0.0418, | |
| "step": 240000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "eval_loss": 0.03996673598885536, | |
| "eval_runtime": 276.1955, | |
| "eval_samples_per_second": 79.762, | |
| "eval_steps_per_second": 19.942, | |
| "step": 240000 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 3.73909315579765e-05, | |
| "loss": 0.0406, | |
| "step": 250000 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 3.688656882029556e-05, | |
| "loss": 0.041, | |
| "step": 260000 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "eval_loss": 0.040077704936265945, | |
| "eval_runtime": 288.7223, | |
| "eval_samples_per_second": 76.302, | |
| "eval_steps_per_second": 19.077, | |
| "step": 260000 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 3.6382206082614615e-05, | |
| "loss": 0.0413, | |
| "step": 270000 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 3.587784334493368e-05, | |
| "loss": 0.0425, | |
| "step": 280000 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "eval_loss": 0.03996116667985916, | |
| "eval_runtime": 267.335, | |
| "eval_samples_per_second": 82.406, | |
| "eval_steps_per_second": 20.603, | |
| "step": 280000 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 3.537348060725274e-05, | |
| "loss": 0.0401, | |
| "step": 290000 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 3.48691178695718e-05, | |
| "loss": 0.0396, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "eval_loss": 0.040498603135347366, | |
| "eval_runtime": 265.1951, | |
| "eval_samples_per_second": 83.071, | |
| "eval_steps_per_second": 20.77, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 3.436475513189086e-05, | |
| "loss": 0.0409, | |
| "step": 310000 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 3.3860392394209916e-05, | |
| "loss": 0.0409, | |
| "step": 320000 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "eval_loss": 0.03931990638375282, | |
| "eval_runtime": 283.7794, | |
| "eval_samples_per_second": 77.631, | |
| "eval_steps_per_second": 19.409, | |
| "step": 320000 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 3.335602965652897e-05, | |
| "loss": 0.0406, | |
| "step": 330000 | |
| } | |
| ], | |
| "max_steps": 991350, | |
| "num_train_epochs": 10, | |
| "total_flos": 2.411321596919808e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |