| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 39.89333333333333, | |
| "eval_steps": 500, | |
| "global_step": 7480, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005333333333333333, | |
| "grad_norm": 3.9845798015594482, | |
| "learning_rate": 1.3368983957219251e-07, | |
| "loss": 1.2155, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.9973333333333333, | |
| "eval_loss": 0.6746753454208374, | |
| "eval_runtime": 12.9251, | |
| "eval_samples_per_second": 86.963, | |
| "eval_steps_per_second": 14.545, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.7129796147346497, | |
| "eval_runtime": 12.5123, | |
| "eval_samples_per_second": 89.831, | |
| "eval_steps_per_second": 15.025, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.997333333333333, | |
| "eval_loss": 0.8005833029747009, | |
| "eval_runtime": 12.5146, | |
| "eval_samples_per_second": 89.815, | |
| "eval_steps_per_second": 15.022, | |
| "step": 562 | |
| }, | |
| { | |
| "epoch": 3.7333333333333334, | |
| "grad_norm": 0.7766222357749939, | |
| "learning_rate": 9.358288770053476e-05, | |
| "loss": 0.582, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.8906850814819336, | |
| "eval_runtime": 12.5313, | |
| "eval_samples_per_second": 89.695, | |
| "eval_steps_per_second": 15.002, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 4.997333333333334, | |
| "eval_loss": 0.9965974688529968, | |
| "eval_runtime": 12.4874, | |
| "eval_samples_per_second": 90.011, | |
| "eval_steps_per_second": 15.055, | |
| "step": 937 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.1142816543579102, | |
| "eval_runtime": 12.5004, | |
| "eval_samples_per_second": 89.917, | |
| "eval_steps_per_second": 15.039, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 6.997333333333334, | |
| "eval_loss": 1.200727939605713, | |
| "eval_runtime": 12.5034, | |
| "eval_samples_per_second": 89.896, | |
| "eval_steps_per_second": 15.036, | |
| "step": 1312 | |
| }, | |
| { | |
| "epoch": 7.466666666666667, | |
| "grad_norm": 0.6246238946914673, | |
| "learning_rate": 0.0001871657754010695, | |
| "loss": 0.2905, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.2675864696502686, | |
| "eval_runtime": 12.484, | |
| "eval_samples_per_second": 90.035, | |
| "eval_steps_per_second": 15.059, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 8.997333333333334, | |
| "eval_loss": 1.3635159730911255, | |
| "eval_runtime": 12.4913, | |
| "eval_samples_per_second": 89.983, | |
| "eval_steps_per_second": 15.05, | |
| "step": 1687 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.4521616697311401, | |
| "eval_runtime": 12.5047, | |
| "eval_samples_per_second": 89.886, | |
| "eval_steps_per_second": 15.034, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 10.997333333333334, | |
| "eval_loss": 1.4860386848449707, | |
| "eval_runtime": 12.4767, | |
| "eval_samples_per_second": 90.088, | |
| "eval_steps_per_second": 15.068, | |
| "step": 2062 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "grad_norm": 0.3173273503780365, | |
| "learning_rate": 0.00019501438995110992, | |
| "loss": 0.1313, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.5470738410949707, | |
| "eval_runtime": 12.4572, | |
| "eval_samples_per_second": 90.229, | |
| "eval_steps_per_second": 15.092, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 12.997333333333334, | |
| "eval_loss": 1.5745112895965576, | |
| "eval_runtime": 12.4452, | |
| "eval_samples_per_second": 90.316, | |
| "eval_steps_per_second": 15.106, | |
| "step": 2437 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 1.6154626607894897, | |
| "eval_runtime": 12.4425, | |
| "eval_samples_per_second": 90.335, | |
| "eval_steps_per_second": 15.109, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 14.933333333333334, | |
| "grad_norm": 0.269999235868454, | |
| "learning_rate": 0.000177467306667847, | |
| "loss": 0.0448, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 14.997333333333334, | |
| "eval_loss": 1.6521568298339844, | |
| "eval_runtime": 12.453, | |
| "eval_samples_per_second": 90.26, | |
| "eval_steps_per_second": 15.097, | |
| "step": 2812 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 1.693098783493042, | |
| "eval_runtime": 12.4351, | |
| "eval_samples_per_second": 90.39, | |
| "eval_steps_per_second": 15.119, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 16.997333333333334, | |
| "eval_loss": 1.7404099702835083, | |
| "eval_runtime": 12.4667, | |
| "eval_samples_per_second": 90.16, | |
| "eval_steps_per_second": 15.08, | |
| "step": 3187 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 1.7414445877075195, | |
| "eval_runtime": 12.4548, | |
| "eval_samples_per_second": 90.246, | |
| "eval_steps_per_second": 15.095, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 18.666666666666668, | |
| "grad_norm": 0.16562360525131226, | |
| "learning_rate": 0.0001495750499965753, | |
| "loss": 0.0268, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 18.997333333333334, | |
| "eval_loss": 1.7563754320144653, | |
| "eval_runtime": 12.9626, | |
| "eval_samples_per_second": 86.711, | |
| "eval_steps_per_second": 14.503, | |
| "step": 3562 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 1.768137812614441, | |
| "eval_runtime": 12.4413, | |
| "eval_samples_per_second": 90.344, | |
| "eval_steps_per_second": 15.111, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 20.997333333333334, | |
| "eval_loss": 1.7924139499664307, | |
| "eval_runtime": 12.441, | |
| "eval_samples_per_second": 90.346, | |
| "eval_steps_per_second": 15.111, | |
| "step": 3937 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 1.832397222518921, | |
| "eval_runtime": 12.4418, | |
| "eval_samples_per_second": 90.34, | |
| "eval_steps_per_second": 15.11, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 22.4, | |
| "grad_norm": 0.10293902456760406, | |
| "learning_rate": 0.0001150624200350232, | |
| "loss": 0.0157, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 22.997333333333334, | |
| "eval_loss": 1.8358964920043945, | |
| "eval_runtime": 12.424, | |
| "eval_samples_per_second": 90.47, | |
| "eval_steps_per_second": 15.132, | |
| "step": 4312 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 1.8424901962280273, | |
| "eval_runtime": 12.4549, | |
| "eval_samples_per_second": 90.246, | |
| "eval_steps_per_second": 15.094, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 24.997333333333334, | |
| "eval_loss": 1.8898138999938965, | |
| "eval_runtime": 12.4368, | |
| "eval_samples_per_second": 90.377, | |
| "eval_steps_per_second": 15.116, | |
| "step": 4687 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 1.9125876426696777, | |
| "eval_runtime": 12.4412, | |
| "eval_samples_per_second": 90.345, | |
| "eval_steps_per_second": 15.111, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 26.133333333333333, | |
| "grad_norm": 0.07291631400585175, | |
| "learning_rate": 7.853831770392755e-05, | |
| "loss": 0.0061, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 26.997333333333334, | |
| "eval_loss": 1.9254367351531982, | |
| "eval_runtime": 12.4413, | |
| "eval_samples_per_second": 90.345, | |
| "eval_steps_per_second": 15.111, | |
| "step": 5062 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 1.975632667541504, | |
| "eval_runtime": 12.4478, | |
| "eval_samples_per_second": 90.297, | |
| "eval_steps_per_second": 15.103, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 28.997333333333334, | |
| "eval_loss": 1.9888043403625488, | |
| "eval_runtime": 12.4518, | |
| "eval_samples_per_second": 90.268, | |
| "eval_steps_per_second": 15.098, | |
| "step": 5437 | |
| }, | |
| { | |
| "epoch": 29.866666666666667, | |
| "grad_norm": 0.000998605857603252, | |
| "learning_rate": 4.4880260858357746e-05, | |
| "loss": 0.0011, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 2.0155458450317383, | |
| "eval_runtime": 12.4346, | |
| "eval_samples_per_second": 90.393, | |
| "eval_steps_per_second": 15.119, | |
| "step": 5625 | |
| }, | |
| { | |
| "epoch": 30.997333333333334, | |
| "eval_loss": 2.048884153366089, | |
| "eval_runtime": 12.435, | |
| "eval_samples_per_second": 90.39, | |
| "eval_steps_per_second": 15.119, | |
| "step": 5812 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 2.070535898208618, | |
| "eval_runtime": 12.426, | |
| "eval_samples_per_second": 90.455, | |
| "eval_steps_per_second": 15.13, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 32.99733333333333, | |
| "eval_loss": 2.087083339691162, | |
| "eval_runtime": 12.428, | |
| "eval_samples_per_second": 90.441, | |
| "eval_steps_per_second": 15.127, | |
| "step": 6187 | |
| }, | |
| { | |
| "epoch": 33.6, | |
| "grad_norm": 0.0002252462727483362, | |
| "learning_rate": 1.8583028637387568e-05, | |
| "loss": 0.0001, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_loss": 2.099353551864624, | |
| "eval_runtime": 12.4444, | |
| "eval_samples_per_second": 90.322, | |
| "eval_steps_per_second": 15.107, | |
| "step": 6375 | |
| }, | |
| { | |
| "epoch": 34.99733333333333, | |
| "eval_loss": 2.1083507537841797, | |
| "eval_runtime": 12.431, | |
| "eval_samples_per_second": 90.419, | |
| "eval_steps_per_second": 15.123, | |
| "step": 6562 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_loss": 2.114396333694458, | |
| "eval_runtime": 12.8335, | |
| "eval_samples_per_second": 87.583, | |
| "eval_steps_per_second": 14.649, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 36.99733333333333, | |
| "eval_loss": 2.1184046268463135, | |
| "eval_runtime": 12.4403, | |
| "eval_samples_per_second": 90.352, | |
| "eval_steps_per_second": 15.112, | |
| "step": 6937 | |
| }, | |
| { | |
| "epoch": 37.333333333333336, | |
| "grad_norm": 0.00023204727040138096, | |
| "learning_rate": 3.1584176769466346e-06, | |
| "loss": 0.0, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_loss": 2.12043833732605, | |
| "eval_runtime": 12.6027, | |
| "eval_samples_per_second": 89.187, | |
| "eval_steps_per_second": 14.917, | |
| "step": 7125 | |
| }, | |
| { | |
| "epoch": 38.99733333333333, | |
| "eval_loss": 2.121260643005371, | |
| "eval_runtime": 12.4106, | |
| "eval_samples_per_second": 90.568, | |
| "eval_steps_per_second": 15.148, | |
| "step": 7312 | |
| }, | |
| { | |
| "epoch": 39.89333333333333, | |
| "eval_loss": 2.1213064193725586, | |
| "eval_runtime": 12.42, | |
| "eval_samples_per_second": 90.499, | |
| "eval_steps_per_second": 15.137, | |
| "step": 7480 | |
| }, | |
| { | |
| "epoch": 39.89333333333333, | |
| "step": 7480, | |
| "total_flos": 1.743333613371392e+18, | |
| "train_loss": 0.10288844243324496, | |
| "train_runtime": 9416.1259, | |
| "train_samples_per_second": 19.108, | |
| "train_steps_per_second": 0.794 | |
| } | |
| ], | |
| "logging_steps": 700, | |
| "max_steps": 7480, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.743333613371392e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |