| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.02564787603526583, | |
| "eval_steps": 8, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0008549292011755277, | |
| "grad_norm": 16.207183837890625, | |
| "learning_rate": 1e-05, | |
| "loss": 9.2165, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0008549292011755277, | |
| "eval_loss": 9.122929573059082, | |
| "eval_runtime": 80.1509, | |
| "eval_samples_per_second": 24.579, | |
| "eval_steps_per_second": 12.289, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0017098584023510553, | |
| "grad_norm": 17.143712997436523, | |
| "learning_rate": 2e-05, | |
| "loss": 8.7316, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.002564787603526583, | |
| "grad_norm": 17.279664993286133, | |
| "learning_rate": 3e-05, | |
| "loss": 8.6895, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0034197168047021107, | |
| "grad_norm": 16.366716384887695, | |
| "learning_rate": 4e-05, | |
| "loss": 8.8251, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.004274646005877639, | |
| "grad_norm": 17.285783767700195, | |
| "learning_rate": 5e-05, | |
| "loss": 8.9384, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.005129575207053166, | |
| "grad_norm": 15.761588096618652, | |
| "learning_rate": 6e-05, | |
| "loss": 8.3716, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.005984504408228694, | |
| "grad_norm": 15.229612350463867, | |
| "learning_rate": 7e-05, | |
| "loss": 7.1605, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.006839433609404221, | |
| "grad_norm": 13.926135063171387, | |
| "learning_rate": 8e-05, | |
| "loss": 6.1561, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.006839433609404221, | |
| "eval_loss": 5.026383876800537, | |
| "eval_runtime": 69.7028, | |
| "eval_samples_per_second": 28.263, | |
| "eval_steps_per_second": 14.131, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.007694362810579749, | |
| "grad_norm": 11.57668399810791, | |
| "learning_rate": 9e-05, | |
| "loss": 5.4513, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.008549292011755277, | |
| "grad_norm": 10.697927474975586, | |
| "learning_rate": 0.0001, | |
| "loss": 4.3594, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.009404221212930804, | |
| "grad_norm": 10.82197093963623, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 3.4726, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.010259150414106332, | |
| "grad_norm": 11.357038497924805, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 2.6108, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.011114079615281859, | |
| "grad_norm": 11.37375259399414, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 1.9079, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.011969008816457388, | |
| "grad_norm": 9.203908920288086, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.1246, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.012823938017632914, | |
| "grad_norm": 5.906091690063477, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.7104, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.013678867218808443, | |
| "grad_norm": 10.134968757629395, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.707, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.013678867218808443, | |
| "eval_loss": 0.6931764483451843, | |
| "eval_runtime": 69.7735, | |
| "eval_samples_per_second": 28.234, | |
| "eval_steps_per_second": 14.117, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.01453379641998397, | |
| "grad_norm": 12.619002342224121, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.5653, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.015388725621159498, | |
| "grad_norm": 8.323529243469238, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.5732, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.016243654822335026, | |
| "grad_norm": 6.280013084411621, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 0.4411, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.017098584023510555, | |
| "grad_norm": 8.90284252166748, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4957, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.01795351322468608, | |
| "grad_norm": 14.239660263061523, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 0.5933, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.018808442425861608, | |
| "grad_norm": 6.530233860015869, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.4186, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.019663371627037136, | |
| "grad_norm": 4.9676833152771, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.412, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.020518300828212665, | |
| "grad_norm": 9.296878814697266, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.497, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.020518300828212665, | |
| "eval_loss": 0.4413914680480957, | |
| "eval_runtime": 70.1243, | |
| "eval_samples_per_second": 28.093, | |
| "eval_steps_per_second": 14.046, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.02137323002938819, | |
| "grad_norm": 3.553572416305542, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.3512, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.022228159230563718, | |
| "grad_norm": 7.732612133026123, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.4699, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.023083088431739247, | |
| "grad_norm": 6.753180503845215, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 0.4656, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.023938017632914775, | |
| "grad_norm": 4.395815372467041, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.4115, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.024792946834090304, | |
| "grad_norm": 6.47470235824585, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.3571, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.02564787603526583, | |
| "grad_norm": 3.6524059772491455, | |
| "learning_rate": 0.0, | |
| "loss": 0.4213, | |
| "step": 30 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 8, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.322534166953984e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |