| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.3489640130861505, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.017448200654307525, | |
| "grad_norm": 14.299415588378906, | |
| "learning_rate": 1e-05, | |
| "loss": 34.4662, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.017448200654307525, | |
| "eval_loss": NaN, | |
| "eval_runtime": 213.7657, | |
| "eval_samples_per_second": 0.454, | |
| "eval_steps_per_second": 0.229, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03489640130861505, | |
| "grad_norm": 14.985342979431152, | |
| "learning_rate": 2e-05, | |
| "loss": 33.5328, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.05234460196292257, | |
| "grad_norm": 12.153814315795898, | |
| "learning_rate": 3e-05, | |
| "loss": 31.8217, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0697928026172301, | |
| "grad_norm": 12.257893562316895, | |
| "learning_rate": 4e-05, | |
| "loss": 31.6887, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.08724100327153762, | |
| "grad_norm": 13.71487808227539, | |
| "learning_rate": 5e-05, | |
| "loss": 34.6487, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.08724100327153762, | |
| "eval_loss": NaN, | |
| "eval_runtime": 2.2625, | |
| "eval_samples_per_second": 42.872, | |
| "eval_steps_per_second": 21.657, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.10468920392584515, | |
| "grad_norm": 14.639806747436523, | |
| "learning_rate": 6e-05, | |
| "loss": 30.0739, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.12213740458015267, | |
| "grad_norm": 17.775012969970703, | |
| "learning_rate": 7e-05, | |
| "loss": 34.9073, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.1395856052344602, | |
| "grad_norm": 18.824005126953125, | |
| "learning_rate": 8e-05, | |
| "loss": 28.7388, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.15703380588876772, | |
| "grad_norm": 23.813568115234375, | |
| "learning_rate": 9e-05, | |
| "loss": 32.9446, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.17448200654307525, | |
| "grad_norm": 26.320598602294922, | |
| "learning_rate": 0.0001, | |
| "loss": 27.4221, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.17448200654307525, | |
| "eval_loss": NaN, | |
| "eval_runtime": 2.2646, | |
| "eval_samples_per_second": 42.833, | |
| "eval_steps_per_second": 21.637, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.19193020719738277, | |
| "grad_norm": 23.103731155395508, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 24.2644, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.2093784078516903, | |
| "grad_norm": 21.649749755859375, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 22.2227, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.2268266085059978, | |
| "grad_norm": 20.080707550048828, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 17.443, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.24427480916030533, | |
| "grad_norm": 17.920568466186523, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 24.8015, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.2617230098146129, | |
| "grad_norm": 17.215288162231445, | |
| "learning_rate": 5e-05, | |
| "loss": 20.5442, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2617230098146129, | |
| "eval_loss": NaN, | |
| "eval_runtime": 2.3117, | |
| "eval_samples_per_second": 41.96, | |
| "eval_steps_per_second": 21.196, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2791712104689204, | |
| "grad_norm": 20.12640380859375, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 23.8817, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.2966194111232279, | |
| "grad_norm": 19.669471740722656, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 14.4217, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.31406761177753545, | |
| "grad_norm": 27.434799194335938, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 17.2192, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.33151581243184297, | |
| "grad_norm": 24.257400512695312, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 15.3634, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.3489640130861505, | |
| "grad_norm": 18.00651741027832, | |
| "learning_rate": 0.0, | |
| "loss": 19.8399, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3489640130861505, | |
| "eval_loss": NaN, | |
| "eval_runtime": 2.2636, | |
| "eval_samples_per_second": 42.852, | |
| "eval_steps_per_second": 21.647, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3152150129475584e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |