| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.044958974935371473, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00044958974935371473, | |
| "eval_loss": 2.36696195602417, | |
| "eval_runtime": 367.0186, | |
| "eval_samples_per_second": 10.209, | |
| "eval_steps_per_second": 1.278, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0013487692480611442, | |
| "grad_norm": 21.040267944335938, | |
| "learning_rate": 3e-05, | |
| "loss": 9.3699, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0026975384961222884, | |
| "grad_norm": 26.55940818786621, | |
| "learning_rate": 6e-05, | |
| "loss": 9.5217, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.004046307744183433, | |
| "grad_norm": 98.55474853515625, | |
| "learning_rate": 9e-05, | |
| "loss": 8.5665, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.004046307744183433, | |
| "eval_loss": 2.109316110610962, | |
| "eval_runtime": 369.6319, | |
| "eval_samples_per_second": 10.137, | |
| "eval_steps_per_second": 1.269, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.005395076992244577, | |
| "grad_norm": 6.886386871337891, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 8.2046, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.006743846240305721, | |
| "grad_norm": 4.885524272918701, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 8.0238, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.008092615488366865, | |
| "grad_norm": 5.15069580078125, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 7.5582, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.008092615488366865, | |
| "eval_loss": 1.8677637577056885, | |
| "eval_runtime": 369.696, | |
| "eval_samples_per_second": 10.135, | |
| "eval_steps_per_second": 1.269, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.00944138473642801, | |
| "grad_norm": 4.313238620758057, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 7.3017, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.010790153984489154, | |
| "grad_norm": 3.7301025390625, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 7.5724, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.012138923232550298, | |
| "grad_norm": 3.6153056621551514, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 7.1896, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.012138923232550298, | |
| "eval_loss": 1.8205772638320923, | |
| "eval_runtime": 369.7914, | |
| "eval_samples_per_second": 10.133, | |
| "eval_steps_per_second": 1.268, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.013487692480611442, | |
| "grad_norm": 3.73301362991333, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 7.4779, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.014836461728672586, | |
| "grad_norm": 3.9209611415863037, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 7.409, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.01618523097673373, | |
| "grad_norm": 4.179993152618408, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 7.2678, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.01618523097673373, | |
| "eval_loss": 1.8027839660644531, | |
| "eval_runtime": 369.7392, | |
| "eval_samples_per_second": 10.134, | |
| "eval_steps_per_second": 1.268, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.017534000224794875, | |
| "grad_norm": 2.9369146823883057, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 7.1185, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.01888276947285602, | |
| "grad_norm": 3.4298548698425293, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 7.395, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.020231538720917163, | |
| "grad_norm": 3.0947425365448, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 6.9911, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.020231538720917163, | |
| "eval_loss": 1.794851541519165, | |
| "eval_runtime": 369.7107, | |
| "eval_samples_per_second": 10.135, | |
| "eval_steps_per_second": 1.269, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.021580307968978307, | |
| "grad_norm": 3.272010326385498, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 7.2596, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.02292907721703945, | |
| "grad_norm": 3.523634195327759, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 7.1694, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.024277846465100596, | |
| "grad_norm": 2.7901206016540527, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 7.1942, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.024277846465100596, | |
| "eval_loss": 1.787168264389038, | |
| "eval_runtime": 369.6887, | |
| "eval_samples_per_second": 10.136, | |
| "eval_steps_per_second": 1.269, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.02562661571316174, | |
| "grad_norm": 3.1662557125091553, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 7.4245, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.026975384961222884, | |
| "grad_norm": 3.515867233276367, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 7.4255, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.02832415420928403, | |
| "grad_norm": 2.9129137992858887, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 7.1692, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.02832415420928403, | |
| "eval_loss": 1.782926082611084, | |
| "eval_runtime": 369.7822, | |
| "eval_samples_per_second": 10.133, | |
| "eval_steps_per_second": 1.268, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.029672923457345173, | |
| "grad_norm": 2.401154041290283, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 7.2576, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.031021692705406317, | |
| "grad_norm": 8.043413162231445, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 7.1562, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.03237046195346746, | |
| "grad_norm": 9.186938285827637, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 6.9678, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.03237046195346746, | |
| "eval_loss": 1.7795237302780151, | |
| "eval_runtime": 369.8817, | |
| "eval_samples_per_second": 10.13, | |
| "eval_steps_per_second": 1.268, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.03371923120152861, | |
| "grad_norm": 6.54693078994751, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 7.0502, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.03506800044958975, | |
| "grad_norm": 3.6234800815582275, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 7.1019, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.0364167696976509, | |
| "grad_norm": 2.8791348934173584, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 7.0355, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.0364167696976509, | |
| "eval_loss": 1.7774608135223389, | |
| "eval_runtime": 369.8032, | |
| "eval_samples_per_second": 10.132, | |
| "eval_steps_per_second": 1.268, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.03776553894571204, | |
| "grad_norm": 3.1315290927886963, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 7.1733, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.039114308193773185, | |
| "grad_norm": 3.2087924480438232, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 7.1793, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.040463077441834326, | |
| "grad_norm": 3.4287378787994385, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 6.9864, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.040463077441834326, | |
| "eval_loss": 1.7765320539474487, | |
| "eval_runtime": 369.8484, | |
| "eval_samples_per_second": 10.131, | |
| "eval_steps_per_second": 1.268, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.041811846689895474, | |
| "grad_norm": 2.9989874362945557, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 7.0778, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.043160615937956615, | |
| "grad_norm": 2.469376802444458, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 6.9031, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.04450938518601776, | |
| "grad_norm": 3.6569230556488037, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 7.0507, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.04450938518601776, | |
| "eval_loss": 1.7763882875442505, | |
| "eval_runtime": 369.9458, | |
| "eval_samples_per_second": 10.129, | |
| "eval_steps_per_second": 1.268, | |
| "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": 2.0531393118968218e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |