| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9933184855233854, | |
| "eval_steps": 100, | |
| "global_step": 294, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.10181355392936685, | |
| "grad_norm": 7.969674587249756, | |
| "learning_rate": 1e-05, | |
| "loss": 3.0773, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2036271078587337, | |
| "grad_norm": 5.803192138671875, | |
| "learning_rate": 2e-05, | |
| "loss": 2.8134, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.30544066178810053, | |
| "grad_norm": 4.133590221405029, | |
| "learning_rate": 3e-05, | |
| "loss": 2.5663, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4072542157174674, | |
| "grad_norm": 0.8438643217086792, | |
| "learning_rate": 4e-05, | |
| "loss": 2.0167, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5090677696468342, | |
| "grad_norm": 0.6160888671875, | |
| "learning_rate": 5e-05, | |
| "loss": 1.9501, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6108813235762011, | |
| "grad_norm": 0.48306068778038025, | |
| "learning_rate": 6e-05, | |
| "loss": 1.9072, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7126948775055679, | |
| "grad_norm": 0.4817599654197693, | |
| "learning_rate": 7e-05, | |
| "loss": 1.8151, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8145084314349348, | |
| "grad_norm": 0.5117820501327515, | |
| "learning_rate": 8e-05, | |
| "loss": 1.8332, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9163219853643016, | |
| "grad_norm": 0.5870208740234375, | |
| "learning_rate": 9e-05, | |
| "loss": 1.8128, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0181355392936684, | |
| "grad_norm": 0.5354984402656555, | |
| "learning_rate": 0.0001, | |
| "loss": 1.9072, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0181355392936684, | |
| "eval_loss": 1.6765505075454712, | |
| "eval_runtime": 215.7529, | |
| "eval_samples_per_second": 1.622, | |
| "eval_steps_per_second": 1.622, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.1199490932230354, | |
| "grad_norm": 0.6077488660812378, | |
| "learning_rate": 9.484536082474227e-05, | |
| "loss": 1.7815, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.2217626471524021, | |
| "grad_norm": 0.6680741906166077, | |
| "learning_rate": 8.969072164948454e-05, | |
| "loss": 1.6772, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.323576201081769, | |
| "grad_norm": 0.6348243355751038, | |
| "learning_rate": 8.453608247422681e-05, | |
| "loss": 1.7182, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.4253897550111359, | |
| "grad_norm": 0.6397783160209656, | |
| "learning_rate": 7.938144329896907e-05, | |
| "loss": 1.7319, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.5272033089405026, | |
| "grad_norm": 0.6410454511642456, | |
| "learning_rate": 7.422680412371135e-05, | |
| "loss": 1.7373, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.6290168628698696, | |
| "grad_norm": 0.6685994863510132, | |
| "learning_rate": 6.907216494845361e-05, | |
| "loss": 1.7746, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.7308304167992365, | |
| "grad_norm": 0.680564284324646, | |
| "learning_rate": 6.391752577319587e-05, | |
| "loss": 1.7363, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.8326439707286033, | |
| "grad_norm": 0.7003260254859924, | |
| "learning_rate": 5.876288659793815e-05, | |
| "loss": 1.6904, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.93445752465797, | |
| "grad_norm": 0.7016832232475281, | |
| "learning_rate": 5.360824742268041e-05, | |
| "loss": 1.6566, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.036271078587337, | |
| "grad_norm": 0.7320058941841125, | |
| "learning_rate": 4.845360824742268e-05, | |
| "loss": 1.8043, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.036271078587337, | |
| "eval_loss": 1.614027500152588, | |
| "eval_runtime": 215.7027, | |
| "eval_samples_per_second": 1.623, | |
| "eval_steps_per_second": 1.623, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.138084632516704, | |
| "grad_norm": 0.7479106783866882, | |
| "learning_rate": 4.329896907216495e-05, | |
| "loss": 1.6897, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.2398981864460707, | |
| "grad_norm": 0.7098692655563354, | |
| "learning_rate": 3.8144329896907216e-05, | |
| "loss": 1.6457, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.3417117403754375, | |
| "grad_norm": 0.7282701134681702, | |
| "learning_rate": 3.2989690721649485e-05, | |
| "loss": 1.7046, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.4435252943048043, | |
| "grad_norm": 0.706913948059082, | |
| "learning_rate": 2.7835051546391755e-05, | |
| "loss": 1.68, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.545338848234171, | |
| "grad_norm": 0.789571225643158, | |
| "learning_rate": 2.268041237113402e-05, | |
| "loss": 1.6881, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.647152402163538, | |
| "grad_norm": 0.7536965012550354, | |
| "learning_rate": 1.7525773195876288e-05, | |
| "loss": 1.6628, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.748965956092905, | |
| "grad_norm": 0.7617937326431274, | |
| "learning_rate": 1.2371134020618558e-05, | |
| "loss": 1.6595, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.8507795100222717, | |
| "grad_norm": 0.7684347629547119, | |
| "learning_rate": 7.216494845360824e-06, | |
| "loss": 1.6476, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.9525930639516385, | |
| "grad_norm": 0.7330012321472168, | |
| "learning_rate": 2.061855670103093e-06, | |
| "loss": 1.6293, | |
| "step": 290 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 294, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.169518958921646e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |