| { | |
| "best_metric": 0.4119344651699066, | |
| "best_model_checkpoint": "/kaggle/working/results/checkpoint-1000", | |
| "epoch": 0.43306707727841404, | |
| "eval_steps": 1000, | |
| "global_step": 9000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.024059282071023, | |
| "grad_norm": 2.836960792541504, | |
| "learning_rate": 3.849485131363681e-06, | |
| "loss": 0.3677, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.048118564142046, | |
| "grad_norm": 3.160072088241577, | |
| "learning_rate": 7.698970262727362e-06, | |
| "loss": 0.386, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.048118564142046, | |
| "eval_loss": 0.4119344651699066, | |
| "eval_runtime": 244.2084, | |
| "eval_samples_per_second": 107.085, | |
| "eval_steps_per_second": 0.839, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.072177846213069, | |
| "grad_norm": 3.1776785850524902, | |
| "learning_rate": 1.154845539409104e-05, | |
| "loss": 0.3816, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.096237128284092, | |
| "grad_norm": 4.521239757537842, | |
| "learning_rate": 1.5397940525454724e-05, | |
| "loss": 0.3767, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.096237128284092, | |
| "eval_loss": 0.4138568937778473, | |
| "eval_runtime": 244.0772, | |
| "eval_samples_per_second": 107.142, | |
| "eval_steps_per_second": 0.84, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.120296410355115, | |
| "grad_norm": 2.133201837539673, | |
| "learning_rate": 1.92474256568184e-05, | |
| "loss": 0.3946, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.144355692426138, | |
| "grad_norm": 3.307325839996338, | |
| "learning_rate": 2.309691078818208e-05, | |
| "loss": 0.3855, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.144355692426138, | |
| "eval_loss": 0.4270198941230774, | |
| "eval_runtime": 244.1676, | |
| "eval_samples_per_second": 107.103, | |
| "eval_steps_per_second": 0.84, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.168414974497161, | |
| "grad_norm": 4.1545729637146, | |
| "learning_rate": 2.6946395919545764e-05, | |
| "loss": 0.3937, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.192474256568184, | |
| "grad_norm": 4.33947229385376, | |
| "learning_rate": 3.079588105090945e-05, | |
| "loss": 0.3935, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.192474256568184, | |
| "eval_loss": 0.42934510111808777, | |
| "eval_runtime": 244.2943, | |
| "eval_samples_per_second": 107.047, | |
| "eval_steps_per_second": 0.839, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.21653353863920702, | |
| "grad_norm": 3.8154330253601074, | |
| "learning_rate": 3.464536618227313e-05, | |
| "loss": 0.3951, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.24059282071023, | |
| "grad_norm": 3.9274327754974365, | |
| "learning_rate": 3.84948513136368e-05, | |
| "loss": 0.3961, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.24059282071023, | |
| "eval_loss": 0.4439002275466919, | |
| "eval_runtime": 244.3634, | |
| "eval_samples_per_second": 107.017, | |
| "eval_steps_per_second": 0.839, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.264652102781253, | |
| "grad_norm": 8.764209747314453, | |
| "learning_rate": 4.234433644500049e-05, | |
| "loss": 0.429, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.288711384852276, | |
| "grad_norm": 2.4573419094085693, | |
| "learning_rate": 4.619382157636416e-05, | |
| "loss": 0.4191, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.288711384852276, | |
| "eval_loss": 0.44426167011260986, | |
| "eval_runtime": 244.015, | |
| "eval_samples_per_second": 107.17, | |
| "eval_steps_per_second": 0.84, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.312770666923299, | |
| "grad_norm": 2.9123361110687256, | |
| "learning_rate": 5.004330670772785e-05, | |
| "loss": 0.4328, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.336829948994322, | |
| "grad_norm": 3.0958211421966553, | |
| "learning_rate": 5.389279183909153e-05, | |
| "loss": 0.4352, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.336829948994322, | |
| "eval_loss": 0.4413658678531647, | |
| "eval_runtime": 244.7271, | |
| "eval_samples_per_second": 106.858, | |
| "eval_steps_per_second": 0.838, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.360889231065345, | |
| "grad_norm": 6.3363037109375, | |
| "learning_rate": 5.77422769704552e-05, | |
| "loss": 0.438, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.384948513136368, | |
| "grad_norm": 5.0592570304870605, | |
| "learning_rate": 6.15917621018189e-05, | |
| "loss": 0.4428, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.384948513136368, | |
| "eval_loss": 0.4693852961063385, | |
| "eval_runtime": 244.8289, | |
| "eval_samples_per_second": 106.813, | |
| "eval_steps_per_second": 0.837, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.409007795207391, | |
| "grad_norm": 3.6804919242858887, | |
| "learning_rate": 6.544124723318256e-05, | |
| "loss": 0.4665, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.43306707727841404, | |
| "grad_norm": 4.925282001495361, | |
| "learning_rate": 6.929073236454625e-05, | |
| "loss": 0.4483, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.43306707727841404, | |
| "eval_loss": 0.4621144235134125, | |
| "eval_runtime": 244.5245, | |
| "eval_samples_per_second": 106.946, | |
| "eval_steps_per_second": 0.838, | |
| "step": 9000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 20782, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.8166939500544e+16, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |