| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.3436426116838488, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.003436426116838488, | |
| "eval_loss": 1.0753878355026245, | |
| "eval_runtime": 70.516, | |
| "eval_samples_per_second": 6.949, | |
| "eval_steps_per_second": 0.879, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010309278350515464, | |
| "grad_norm": 4.210477828979492, | |
| "learning_rate": 3e-05, | |
| "loss": 4.2729, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.020618556701030927, | |
| "grad_norm": 4.062213897705078, | |
| "learning_rate": 6e-05, | |
| "loss": 4.2202, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.030927835051546393, | |
| "grad_norm": 3.1250791549682617, | |
| "learning_rate": 9e-05, | |
| "loss": 4.0332, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.030927835051546393, | |
| "eval_loss": 0.9257749319076538, | |
| "eval_runtime": 71.652, | |
| "eval_samples_per_second": 6.839, | |
| "eval_steps_per_second": 0.865, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.041237113402061855, | |
| "grad_norm": 3.668790102005005, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 3.3551, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.05154639175257732, | |
| "grad_norm": 2.477992534637451, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 3.4003, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.061855670103092786, | |
| "grad_norm": 2.6562209129333496, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 3.3175, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.061855670103092786, | |
| "eval_loss": 0.7811936736106873, | |
| "eval_runtime": 71.6965, | |
| "eval_samples_per_second": 6.834, | |
| "eval_steps_per_second": 0.865, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.07216494845360824, | |
| "grad_norm": 2.1851630210876465, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 2.9538, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.08247422680412371, | |
| "grad_norm": 2.0499980449676514, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 3.2137, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.09278350515463918, | |
| "grad_norm": 1.8330583572387695, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 2.9288, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.09278350515463918, | |
| "eval_loss": 0.7432103753089905, | |
| "eval_runtime": 71.766, | |
| "eval_samples_per_second": 6.828, | |
| "eval_steps_per_second": 0.864, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.10309278350515463, | |
| "grad_norm": 5.810397624969482, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 2.9585, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1134020618556701, | |
| "grad_norm": 2.008681058883667, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 3.021, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.12371134020618557, | |
| "grad_norm": 1.791285514831543, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 2.7954, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.12371134020618557, | |
| "eval_loss": 0.727917492389679, | |
| "eval_runtime": 71.7499, | |
| "eval_samples_per_second": 6.829, | |
| "eval_steps_per_second": 0.864, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.13402061855670103, | |
| "grad_norm": 1.8717230558395386, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 2.7933, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.14432989690721648, | |
| "grad_norm": 1.8673449754714966, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 2.7106, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.15463917525773196, | |
| "grad_norm": 1.9052813053131104, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 2.8453, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.15463917525773196, | |
| "eval_loss": 0.7204706072807312, | |
| "eval_runtime": 71.6809, | |
| "eval_samples_per_second": 6.836, | |
| "eval_steps_per_second": 0.865, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.16494845360824742, | |
| "grad_norm": 1.9197566509246826, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 3.0862, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.17525773195876287, | |
| "grad_norm": 2.105869770050049, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 2.8316, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.18556701030927836, | |
| "grad_norm": 2.0447919368743896, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 2.7667, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.18556701030927836, | |
| "eval_loss": 0.7139142751693726, | |
| "eval_runtime": 71.716, | |
| "eval_samples_per_second": 6.833, | |
| "eval_steps_per_second": 0.865, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.1958762886597938, | |
| "grad_norm": 1.945418119430542, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 2.9707, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.20618556701030927, | |
| "grad_norm": 2.055152416229248, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 3.0264, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.21649484536082475, | |
| "grad_norm": 1.9689520597457886, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 2.7934, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.21649484536082475, | |
| "eval_loss": 0.7113768458366394, | |
| "eval_runtime": 71.6776, | |
| "eval_samples_per_second": 6.836, | |
| "eval_steps_per_second": 0.865, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.2268041237113402, | |
| "grad_norm": 1.9071241617202759, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 2.8892, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.23711340206185566, | |
| "grad_norm": 2.208230972290039, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 2.9481, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.24742268041237114, | |
| "grad_norm": 1.8062189817428589, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 2.9625, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.24742268041237114, | |
| "eval_loss": 0.7076421976089478, | |
| "eval_runtime": 71.6586, | |
| "eval_samples_per_second": 6.838, | |
| "eval_steps_per_second": 0.865, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.25773195876288657, | |
| "grad_norm": 1.9327573776245117, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 3.193, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.26804123711340205, | |
| "grad_norm": 1.7276326417922974, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 3.0274, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.27835051546391754, | |
| "grad_norm": 1.6160300970077515, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 2.8568, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.27835051546391754, | |
| "eval_loss": 0.7067274451255798, | |
| "eval_runtime": 71.6385, | |
| "eval_samples_per_second": 6.84, | |
| "eval_steps_per_second": 0.865, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.28865979381443296, | |
| "grad_norm": 1.94657301902771, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 2.6903, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.29896907216494845, | |
| "grad_norm": 1.8003250360488892, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 2.9591, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.30927835051546393, | |
| "grad_norm": 1.829463005065918, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 3.063, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.30927835051546393, | |
| "eval_loss": 0.7055235505104065, | |
| "eval_runtime": 71.6799, | |
| "eval_samples_per_second": 6.836, | |
| "eval_steps_per_second": 0.865, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.31958762886597936, | |
| "grad_norm": 1.5607446432113647, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 2.7605, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.32989690721649484, | |
| "grad_norm": 1.7891669273376465, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 2.8084, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.3402061855670103, | |
| "grad_norm": 1.7857033014297485, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 2.8207, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.3402061855670103, | |
| "eval_loss": 0.7051608562469482, | |
| "eval_runtime": 71.6692, | |
| "eval_samples_per_second": 6.837, | |
| "eval_steps_per_second": 0.865, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.5224119259693056e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |