| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.031413612565445025, | |
| "eval_steps": 4, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0010471204188481676, | |
| "grad_norm": 0.4058476686477661, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 3.3086, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0010471204188481676, | |
| "eval_loss": 1.7368364334106445, | |
| "eval_runtime": 50.7078, | |
| "eval_samples_per_second": 7.948, | |
| "eval_steps_per_second": 1.992, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0020942408376963353, | |
| "grad_norm": 0.43642497062683105, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 3.3639, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0031413612565445027, | |
| "grad_norm": 0.49189406633377075, | |
| "learning_rate": 6e-06, | |
| "loss": 3.4682, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.004188481675392671, | |
| "grad_norm": 0.7861902713775635, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 3.727, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.004188481675392671, | |
| "eval_loss": 1.7365564107894897, | |
| "eval_runtime": 50.6919, | |
| "eval_samples_per_second": 7.95, | |
| "eval_steps_per_second": 1.992, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.005235602094240838, | |
| "grad_norm": 0.40293604135513306, | |
| "learning_rate": 1e-05, | |
| "loss": 3.5665, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0062827225130890054, | |
| "grad_norm": 0.4985096752643585, | |
| "learning_rate": 9.960573506572391e-06, | |
| "loss": 3.5222, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.007329842931937173, | |
| "grad_norm": 0.5579627156257629, | |
| "learning_rate": 9.842915805643156e-06, | |
| "loss": 3.4918, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.008376963350785341, | |
| "grad_norm": 0.5257052183151245, | |
| "learning_rate": 9.648882429441258e-06, | |
| "loss": 3.5749, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.008376963350785341, | |
| "eval_loss": 1.7355324029922485, | |
| "eval_runtime": 50.6951, | |
| "eval_samples_per_second": 7.949, | |
| "eval_steps_per_second": 1.992, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.009424083769633508, | |
| "grad_norm": 0.5943679809570312, | |
| "learning_rate": 9.381533400219319e-06, | |
| "loss": 3.4635, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.010471204188481676, | |
| "grad_norm": 0.4801727533340454, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 3.3975, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.011518324607329843, | |
| "grad_norm": 0.7009990811347961, | |
| "learning_rate": 8.644843137107058e-06, | |
| "loss": 3.7071, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.012565445026178011, | |
| "grad_norm": 0.45381149649620056, | |
| "learning_rate": 8.18711994874345e-06, | |
| "loss": 3.4273, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.012565445026178011, | |
| "eval_loss": 1.7340854406356812, | |
| "eval_runtime": 50.7211, | |
| "eval_samples_per_second": 7.945, | |
| "eval_steps_per_second": 1.991, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.013612565445026177, | |
| "grad_norm": 0.7830641269683838, | |
| "learning_rate": 7.679133974894984e-06, | |
| "loss": 3.7301, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.014659685863874346, | |
| "grad_norm": 0.34250831604003906, | |
| "learning_rate": 7.128896457825364e-06, | |
| "loss": 3.0747, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.015706806282722512, | |
| "grad_norm": 0.3523239195346832, | |
| "learning_rate": 6.545084971874738e-06, | |
| "loss": 2.9538, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.016753926701570682, | |
| "grad_norm": 0.5659212470054626, | |
| "learning_rate": 5.936906572928625e-06, | |
| "loss": 3.3864, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.016753926701570682, | |
| "eval_loss": 1.7325068712234497, | |
| "eval_runtime": 50.708, | |
| "eval_samples_per_second": 7.947, | |
| "eval_steps_per_second": 1.992, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.01780104712041885, | |
| "grad_norm": 0.4708373248577118, | |
| "learning_rate": 5.3139525976465675e-06, | |
| "loss": 3.2516, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.018848167539267015, | |
| "grad_norm": 0.5167136192321777, | |
| "learning_rate": 4.686047402353433e-06, | |
| "loss": 3.6126, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.019895287958115182, | |
| "grad_norm": 0.47558876872062683, | |
| "learning_rate": 4.063093427071376e-06, | |
| "loss": 3.3764, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.020942408376963352, | |
| "grad_norm": 0.43575918674468994, | |
| "learning_rate": 3.4549150281252635e-06, | |
| "loss": 2.9594, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.020942408376963352, | |
| "eval_loss": 1.7311694622039795, | |
| "eval_runtime": 50.6622, | |
| "eval_samples_per_second": 7.955, | |
| "eval_steps_per_second": 1.994, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02198952879581152, | |
| "grad_norm": 0.5128723382949829, | |
| "learning_rate": 2.871103542174637e-06, | |
| "loss": 3.1497, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.023036649214659685, | |
| "grad_norm": 0.6108046174049377, | |
| "learning_rate": 2.320866025105016e-06, | |
| "loss": 3.3197, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.024083769633507852, | |
| "grad_norm": 0.45397162437438965, | |
| "learning_rate": 1.8128800512565514e-06, | |
| "loss": 3.2883, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.025130890052356022, | |
| "grad_norm": 0.5446158647537231, | |
| "learning_rate": 1.3551568628929434e-06, | |
| "loss": 3.4563, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.025130890052356022, | |
| "eval_loss": 1.7306746244430542, | |
| "eval_runtime": 50.6678, | |
| "eval_samples_per_second": 7.954, | |
| "eval_steps_per_second": 1.993, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.02617801047120419, | |
| "grad_norm": 0.4350343346595764, | |
| "learning_rate": 9.549150281252633e-07, | |
| "loss": 3.4126, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.027225130890052355, | |
| "grad_norm": 0.49152126908302307, | |
| "learning_rate": 6.184665997806832e-07, | |
| "loss": 3.5553, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.028272251308900525, | |
| "grad_norm": 0.38575446605682373, | |
| "learning_rate": 3.511175705587433e-07, | |
| "loss": 3.2485, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.02931937172774869, | |
| "grad_norm": 0.5538058280944824, | |
| "learning_rate": 1.5708419435684463e-07, | |
| "loss": 3.663, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.02931937172774869, | |
| "eval_loss": 1.7304288148880005, | |
| "eval_runtime": 50.7219, | |
| "eval_samples_per_second": 7.945, | |
| "eval_steps_per_second": 1.991, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.030366492146596858, | |
| "grad_norm": 0.5756243467330933, | |
| "learning_rate": 3.9426493427611177e-08, | |
| "loss": 3.3005, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.031413612565445025, | |
| "grad_norm": 0.4113498628139496, | |
| "learning_rate": 0.0, | |
| "loss": 3.3099, | |
| "step": 30 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 20, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9801969665310720.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |