| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 15.0, | |
| "eval_steps": 10, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.7314285714285714, | |
| "grad_norm": 9.340608453115205, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 0.7205065488815308, | |
| "memory(GiB)": 63.44, | |
| "step": 1, | |
| "token_acc": 0.8149569178102063, | |
| "train_speed(iter/s)": 0.002817 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 9.340608453115205, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.7595161199569702, | |
| "memory(GiB)": 67.12, | |
| "step": 2, | |
| "token_acc": 0.8210892726643333, | |
| "train_speed(iter/s)": 0.004132 | |
| }, | |
| { | |
| "epoch": 1.7314285714285713, | |
| "grad_norm": 14.990541756622823, | |
| "learning_rate": 1.5e-06, | |
| "loss": 0.7181279063224792, | |
| "memory(GiB)": 67.12, | |
| "step": 3, | |
| "token_acc": 0.8385703913117363, | |
| "train_speed(iter/s)": 0.003617 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 14.990541756622823, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.7464776039123535, | |
| "memory(GiB)": 67.12, | |
| "step": 4, | |
| "token_acc": 0.8302431930811472, | |
| "train_speed(iter/s)": 0.004181 | |
| }, | |
| { | |
| "epoch": 2.7314285714285713, | |
| "grad_norm": 10.51872439314909, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.7061335444450378, | |
| "memory(GiB)": 67.12, | |
| "step": 5, | |
| "token_acc": 0.8495207453582679, | |
| "train_speed(iter/s)": 0.003827 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 7.8322768971767545, | |
| "learning_rate": 3e-06, | |
| "loss": 0.7670756578445435, | |
| "memory(GiB)": 76.32, | |
| "step": 6, | |
| "token_acc": 0.8034914159988437, | |
| "train_speed(iter/s)": 0.00417 | |
| }, | |
| { | |
| "epoch": 3.7314285714285713, | |
| "grad_norm": 9.640548264571718, | |
| "learning_rate": 3.5e-06, | |
| "loss": 0.7187657952308655, | |
| "memory(GiB)": 76.32, | |
| "step": 7, | |
| "token_acc": 0.8349510177213882, | |
| "train_speed(iter/s)": 0.003914 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 9.640548264571718, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.6732273101806641, | |
| "memory(GiB)": 76.32, | |
| "step": 8, | |
| "token_acc": 0.8445810474364907, | |
| "train_speed(iter/s)": 0.004178 | |
| }, | |
| { | |
| "epoch": 4.731428571428571, | |
| "grad_norm": 18.16921963489396, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.6695082187652588, | |
| "memory(GiB)": 76.32, | |
| "step": 9, | |
| "token_acc": 0.8329077399296925, | |
| "train_speed(iter/s)": 0.003975 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 18.16921963489396, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7095792293548584, | |
| "memory(GiB)": 76.32, | |
| "step": 10, | |
| "token_acc": 0.8494555815610054, | |
| "train_speed(iter/s)": 0.00417 | |
| }, | |
| { | |
| "epoch": 5.731428571428571, | |
| "grad_norm": 36.47852854409781, | |
| "learning_rate": 4.99847706754774e-06, | |
| "loss": 0.6005474328994751, | |
| "memory(GiB)": 76.32, | |
| "step": 11, | |
| "token_acc": 0.844029136120312, | |
| "train_speed(iter/s)": 0.003971 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 8.15921427581055, | |
| "learning_rate": 4.993910125649561e-06, | |
| "loss": 0.6586757898330688, | |
| "memory(GiB)": 76.32, | |
| "step": 12, | |
| "token_acc": 0.8276634855085236, | |
| "train_speed(iter/s)": 0.004131 | |
| }, | |
| { | |
| "epoch": 6.731428571428571, | |
| "grad_norm": 9.940528313438021, | |
| "learning_rate": 4.986304738420684e-06, | |
| "loss": 0.5984752178192139, | |
| "memory(GiB)": 76.32, | |
| "step": 13, | |
| "token_acc": 0.8406243779216233, | |
| "train_speed(iter/s)": 0.004005 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 9.940528313438021, | |
| "learning_rate": 4.975670171853926e-06, | |
| "loss": 0.5597701668739319, | |
| "memory(GiB)": 76.32, | |
| "step": 14, | |
| "token_acc": 0.8693022202084277, | |
| "train_speed(iter/s)": 0.004144 | |
| }, | |
| { | |
| "epoch": 7.731428571428571, | |
| "grad_norm": 4.5395059579286015, | |
| "learning_rate": 4.962019382530521e-06, | |
| "loss": 0.5627670884132385, | |
| "memory(GiB)": 76.32, | |
| "step": 15, | |
| "token_acc": 0.8462122810719561, | |
| "train_speed(iter/s)": 0.004014 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.5395059579286015, | |
| "learning_rate": 4.9453690018345144e-06, | |
| "loss": 0.5592302680015564, | |
| "memory(GiB)": 76.32, | |
| "step": 16, | |
| "token_acc": 0.8405773019678591, | |
| "train_speed(iter/s)": 0.004144 | |
| }, | |
| { | |
| "epoch": 8.731428571428571, | |
| "grad_norm": 6.584180692482273, | |
| "learning_rate": 4.925739315689991e-06, | |
| "loss": 0.49423277378082275, | |
| "memory(GiB)": 76.32, | |
| "step": 17, | |
| "token_acc": 0.8678222664079662, | |
| "train_speed(iter/s)": 0.004034 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 2.39165864781694, | |
| "learning_rate": 4.903154239845798e-06, | |
| "loss": 0.4701133668422699, | |
| "memory(GiB)": 76.32, | |
| "step": 18, | |
| "token_acc": 0.8865700144198917, | |
| "train_speed(iter/s)": 0.004143 | |
| }, | |
| { | |
| "epoch": 9.731428571428571, | |
| "grad_norm": 5.595725551844695, | |
| "learning_rate": 4.8776412907378845e-06, | |
| "loss": 0.45911359786987305, | |
| "memory(GiB)": 76.32, | |
| "step": 19, | |
| "token_acc": 0.8713313357013871, | |
| "train_speed(iter/s)": 0.004052 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 5.595725551844695, | |
| "learning_rate": 4.849231551964771e-06, | |
| "loss": 0.4861743152141571, | |
| "memory(GiB)": 76.32, | |
| "step": 20, | |
| "token_acc": 0.854624478442281, | |
| "train_speed(iter/s)": 0.00415 | |
| }, | |
| { | |
| "epoch": 10.731428571428571, | |
| "grad_norm": 2.9321408358661594, | |
| "learning_rate": 4.817959636416969e-06, | |
| "loss": 0.45954838395118713, | |
| "memory(GiB)": 76.32, | |
| "step": 21, | |
| "token_acc": 0.8712813064578979, | |
| "train_speed(iter/s)": 0.004039 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 11.05893897846086, | |
| "learning_rate": 4.783863644106502e-06, | |
| "loss": 0.42236876487731934, | |
| "memory(GiB)": 76.32, | |
| "step": 22, | |
| "token_acc": 0.8768863403778366, | |
| "train_speed(iter/s)": 0.004131 | |
| }, | |
| { | |
| "epoch": 11.731428571428571, | |
| "grad_norm": 3.349470799070857, | |
| "learning_rate": 4.746985115747918e-06, | |
| "loss": 0.4331884980201721, | |
| "memory(GiB)": 76.32, | |
| "step": 23, | |
| "token_acc": 0.8758490488600685, | |
| "train_speed(iter/s)": 0.004053 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 3.349470799070857, | |
| "learning_rate": 4.707368982147318e-06, | |
| "loss": 0.4267829954624176, | |
| "memory(GiB)": 76.32, | |
| "step": 24, | |
| "token_acc": 0.8883371910699619, | |
| "train_speed(iter/s)": 0.004139 | |
| }, | |
| { | |
| "epoch": 12.731428571428571, | |
| "grad_norm": 4.894666598752148, | |
| "learning_rate": 4.665063509461098e-06, | |
| "loss": 0.4266759753227234, | |
| "memory(GiB)": 76.32, | |
| "step": 25, | |
| "token_acc": 0.879163815519365, | |
| "train_speed(iter/s)": 0.004064 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 4.894666598752148, | |
| "learning_rate": 4.620120240391065e-06, | |
| "loss": 0.3838977813720703, | |
| "memory(GiB)": 76.32, | |
| "step": 26, | |
| "token_acc": 0.8920969339679625, | |
| "train_speed(iter/s)": 0.004143 | |
| }, | |
| { | |
| "epoch": 13.731428571428571, | |
| "grad_norm": 2.1762524149258344, | |
| "learning_rate": 4.572593931387604e-06, | |
| "loss": 0.3890763521194458, | |
| "memory(GiB)": 76.32, | |
| "step": 27, | |
| "token_acc": 0.8864340359319397, | |
| "train_speed(iter/s)": 0.004074 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 1.8853256148082378, | |
| "learning_rate": 4.522542485937369e-06, | |
| "loss": 0.36992955207824707, | |
| "memory(GiB)": 76.32, | |
| "step": 28, | |
| "token_acc": 0.8742953776775648, | |
| "train_speed(iter/s)": 0.004147 | |
| }, | |
| { | |
| "epoch": 14.731428571428571, | |
| "grad_norm": 1.3310226706467028, | |
| "learning_rate": 4.470026884016805e-06, | |
| "loss": 0.3653033375740051, | |
| "memory(GiB)": 76.32, | |
| "step": 29, | |
| "token_acc": 0.8962616884718178, | |
| "train_speed(iter/s)": 0.004085 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.3310226706467028, | |
| "learning_rate": 4.415111107797445e-06, | |
| "loss": 0.3536423146724701, | |
| "memory(GiB)": 76.32, | |
| "step": 30, | |
| "token_acc": 0.8946604521852897, | |
| "train_speed(iter/s)": 0.004149 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 57394211913728.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |