| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 100.0, | |
| "eval_steps": 390, | |
| "global_step": 3900, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 12.714141845703125, | |
| "learning_rate": 9e-06, | |
| "loss": 3.3102, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 2.768110990524292, | |
| "eval_runtime": 9.8747, | |
| "eval_samples_per_second": 252.057, | |
| "eval_steps_per_second": 2.532, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 13.40042781829834, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.6079, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 1.5404176712036133, | |
| "eval_runtime": 9.8518, | |
| "eval_samples_per_second": 252.644, | |
| "eval_steps_per_second": 2.538, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 11.014426231384277, | |
| "learning_rate": 7e-06, | |
| "loss": 0.7749, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 0.9965859651565552, | |
| "eval_runtime": 9.5463, | |
| "eval_samples_per_second": 260.73, | |
| "eval_steps_per_second": 2.619, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 5.882655620574951, | |
| "learning_rate": 6e-06, | |
| "loss": 0.4468, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 0.7464911937713623, | |
| "eval_runtime": 9.522, | |
| "eval_samples_per_second": 261.395, | |
| "eval_steps_per_second": 2.626, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 6.569639682769775, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2965, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_loss": 0.5969729423522949, | |
| "eval_runtime": 9.4996, | |
| "eval_samples_per_second": 262.01, | |
| "eval_steps_per_second": 2.632, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 4.880886554718018, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.2199, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_loss": 0.5014213919639587, | |
| "eval_runtime": 9.4946, | |
| "eval_samples_per_second": 262.15, | |
| "eval_steps_per_second": 2.633, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "grad_norm": 5.099749565124512, | |
| "learning_rate": 3e-06, | |
| "loss": 0.1751, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_loss": 0.44686585664749146, | |
| "eval_runtime": 9.4777, | |
| "eval_samples_per_second": 262.616, | |
| "eval_steps_per_second": 2.638, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 3.513817548751831, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.1487, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_loss": 0.402406245470047, | |
| "eval_runtime": 9.4812, | |
| "eval_samples_per_second": 262.52, | |
| "eval_steps_per_second": 2.637, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "grad_norm": 2.282205104827881, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.1317, | |
| "step": 3510 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_loss": 0.3745902478694916, | |
| "eval_runtime": 9.5617, | |
| "eval_samples_per_second": 260.309, | |
| "eval_steps_per_second": 2.615, | |
| "step": 3510 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 2.7403271198272705, | |
| "learning_rate": 0.0, | |
| "loss": 0.1234, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_loss": 0.37145158648490906, | |
| "eval_runtime": 9.4722, | |
| "eval_samples_per_second": 262.769, | |
| "eval_steps_per_second": 2.639, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 3900, | |
| "total_flos": 6.5849779842816e+16, | |
| "train_loss": 0.7235021141247872, | |
| "train_runtime": 2314.4669, | |
| "train_samples_per_second": 107.541, | |
| "train_steps_per_second": 1.685 | |
| } | |
| ], | |
| "logging_steps": 390, | |
| "max_steps": 3900, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "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": 6.5849779842816e+16, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |