| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.8849557522123894, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.008849557522123894, | |
| "eval_loss": 7.850747108459473, | |
| "eval_runtime": 11.9245, | |
| "eval_samples_per_second": 7.967, | |
| "eval_steps_per_second": 2.013, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02654867256637168, | |
| "grad_norm": 23.30268096923828, | |
| "learning_rate": 3e-05, | |
| "loss": 32.3052, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.05309734513274336, | |
| "grad_norm": 24.82865333557129, | |
| "learning_rate": 6e-05, | |
| "loss": 26.7374, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.07964601769911504, | |
| "grad_norm": 16.411090850830078, | |
| "learning_rate": 9e-05, | |
| "loss": 15.0507, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.07964601769911504, | |
| "eval_loss": 1.889989972114563, | |
| "eval_runtime": 12.1586, | |
| "eval_samples_per_second": 7.813, | |
| "eval_steps_per_second": 1.974, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.10619469026548672, | |
| "grad_norm": 16.30251693725586, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 4.4943, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.13274336283185842, | |
| "grad_norm": 2.594444513320923, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 1.7033, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1592920353982301, | |
| "grad_norm": 0.005044514779001474, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.0137, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.1592920353982301, | |
| "eval_loss": 0.0003469553776085377, | |
| "eval_runtime": 12.2289, | |
| "eval_samples_per_second": 7.769, | |
| "eval_steps_per_second": 1.963, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.18584070796460178, | |
| "grad_norm": 0.012934673577547073, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.0003, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.21238938053097345, | |
| "grad_norm": 0.00015438912669196725, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.0002, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.23893805309734514, | |
| "grad_norm": 0.0381203256547451, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.0001, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.23893805309734514, | |
| "eval_loss": 1.8848184026865056e-06, | |
| "eval_runtime": 12.2379, | |
| "eval_samples_per_second": 7.763, | |
| "eval_steps_per_second": 1.961, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.26548672566371684, | |
| "grad_norm": 0.005300083197653294, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2920353982300885, | |
| "grad_norm": 0.0005808924324810505, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.0, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.3185840707964602, | |
| "grad_norm": 0.8387057185173035, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.0023, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.3185840707964602, | |
| "eval_loss": 2.120336603184114e-06, | |
| "eval_runtime": 12.2271, | |
| "eval_samples_per_second": 7.77, | |
| "eval_steps_per_second": 1.963, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.34513274336283184, | |
| "grad_norm": 0.001999981701374054, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.0, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.37168141592920356, | |
| "grad_norm": 0.013748248107731342, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.0001, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.39823008849557523, | |
| "grad_norm": 0.003550400724634528, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.0001, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.39823008849557523, | |
| "eval_loss": 2.637667603266891e-05, | |
| "eval_runtime": 12.2233, | |
| "eval_samples_per_second": 7.772, | |
| "eval_steps_per_second": 1.963, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.4247787610619469, | |
| "grad_norm": 0.00013267536996863782, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.0005, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.45132743362831856, | |
| "grad_norm": 0.00011587361223064363, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.0, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.4778761061946903, | |
| "grad_norm": 6.926426431164145e-05, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.0, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.4778761061946903, | |
| "eval_loss": 2.020281613113184e-07, | |
| "eval_runtime": 12.2268, | |
| "eval_samples_per_second": 7.77, | |
| "eval_steps_per_second": 1.963, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.504424778761062, | |
| "grad_norm": 4.16663424402941e-05, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.0, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.5309734513274337, | |
| "grad_norm": 3.846506660920568e-05, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5575221238938053, | |
| "grad_norm": 5.5511947721242905e-05, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.0, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.5575221238938053, | |
| "eval_loss": 2.6978867140314833e-07, | |
| "eval_runtime": 12.2104, | |
| "eval_samples_per_second": 7.78, | |
| "eval_steps_per_second": 1.966, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.584070796460177, | |
| "grad_norm": 3.3250638807658106e-05, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.0, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.6106194690265486, | |
| "grad_norm": 0.0001759525912348181, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.0, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.6371681415929203, | |
| "grad_norm": 0.00013978495553601533, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.0, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.6371681415929203, | |
| "eval_loss": 2.387317579177761e-07, | |
| "eval_runtime": 12.2393, | |
| "eval_samples_per_second": 7.762, | |
| "eval_steps_per_second": 1.961, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.6637168141592921, | |
| "grad_norm": 4.825639189220965e-05, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.0, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.6902654867256637, | |
| "grad_norm": 0.0025355853140354156, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.0, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.7168141592920354, | |
| "grad_norm": 6.508089427370578e-05, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 0.0, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.7168141592920354, | |
| "eval_loss": 2.18654548689301e-07, | |
| "eval_runtime": 12.2613, | |
| "eval_samples_per_second": 7.748, | |
| "eval_steps_per_second": 1.957, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.7433628318584071, | |
| "grad_norm": 6.199401832418516e-05, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.0, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.7699115044247787, | |
| "grad_norm": 5.3589974413625896e-05, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.0, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.7964601769911505, | |
| "grad_norm": 0.003909613471478224, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.7964601769911505, | |
| "eval_loss": 2.0171441406091617e-07, | |
| "eval_runtime": 12.2288, | |
| "eval_samples_per_second": 7.769, | |
| "eval_steps_per_second": 1.963, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8230088495575221, | |
| "grad_norm": 7.868893590057269e-05, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.0, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.8495575221238938, | |
| "grad_norm": 4.243826697347686e-05, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.0, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.8761061946902655, | |
| "grad_norm": 6.109167588874698e-05, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 0.0, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.8761061946902655, | |
| "eval_loss": 2.0830219682466122e-07, | |
| "eval_runtime": 12.2672, | |
| "eval_samples_per_second": 7.744, | |
| "eval_steps_per_second": 1.956, | |
| "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": 1.4586867433340928e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |