| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.37593984962406013, | |
| "eval_steps": 63, | |
| "global_step": 250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0015037593984962407, | |
| "eval_loss": 10.374281883239746, | |
| "eval_runtime": 0.3459, | |
| "eval_samples_per_second": 404.758, | |
| "eval_steps_per_second": 202.379, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.015037593984962405, | |
| "grad_norm": 0.1071339100599289, | |
| "learning_rate": 0.00019979453927503364, | |
| "loss": 10.3714, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03007518796992481, | |
| "grad_norm": 0.13845373690128326, | |
| "learning_rate": 0.00019815591569910654, | |
| "loss": 10.37, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.045112781954887216, | |
| "grad_norm": 0.2508547902107239, | |
| "learning_rate": 0.00019490557470106686, | |
| "loss": 10.3627, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06015037593984962, | |
| "grad_norm": 0.1770472228527069, | |
| "learning_rate": 0.0001900968867902419, | |
| "loss": 10.361, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.07518796992481203, | |
| "grad_norm": 0.13482050597667694, | |
| "learning_rate": 0.00018380881048918405, | |
| "loss": 10.3585, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.09022556390977443, | |
| "grad_norm": 0.13020801544189453, | |
| "learning_rate": 0.00017614459583691346, | |
| "loss": 10.3527, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.09473684210526316, | |
| "eval_loss": 10.353063583374023, | |
| "eval_runtime": 0.3419, | |
| "eval_samples_per_second": 409.491, | |
| "eval_steps_per_second": 204.746, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.10526315789473684, | |
| "grad_norm": 0.14389438927173615, | |
| "learning_rate": 0.0001672300890261317, | |
| "loss": 10.3523, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.12030075187969924, | |
| "grad_norm": 0.1498968005180359, | |
| "learning_rate": 0.00015721166601221698, | |
| "loss": 10.3438, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.13533834586466165, | |
| "grad_norm": 0.20479345321655273, | |
| "learning_rate": 0.00014625382902408356, | |
| "loss": 10.3466, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.15037593984962405, | |
| "grad_norm": 0.19941863417625427, | |
| "learning_rate": 0.00013453650544213076, | |
| "loss": 10.3509, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.16541353383458646, | |
| "grad_norm": 0.19158747792243958, | |
| "learning_rate": 0.00012225209339563145, | |
| "loss": 10.3451, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.18045112781954886, | |
| "grad_norm": 0.15758362412452698, | |
| "learning_rate": 0.00010960230259076818, | |
| "loss": 10.3473, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.18947368421052632, | |
| "eval_loss": 10.345841407775879, | |
| "eval_runtime": 0.3607, | |
| "eval_samples_per_second": 388.186, | |
| "eval_steps_per_second": 194.093, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.19548872180451127, | |
| "grad_norm": 0.15928584337234497, | |
| "learning_rate": 9.679484224283449e-05, | |
| "loss": 10.3503, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.21052631578947367, | |
| "grad_norm": 0.21827955543994904, | |
| "learning_rate": 8.404001049666211e-05, | |
| "loss": 10.3474, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.22556390977443608, | |
| "grad_norm": 0.1820470094680786, | |
| "learning_rate": 7.154724133689677e-05, | |
| "loss": 10.3433, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.24060150375939848, | |
| "grad_norm": 0.19076405465602875, | |
| "learning_rate": 5.952166568776062e-05, | |
| "loss": 10.3443, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.2556390977443609, | |
| "grad_norm": 0.18246634304523468, | |
| "learning_rate": 4.8160743168947496e-05, | |
| "loss": 10.3442, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.2706766917293233, | |
| "grad_norm": 0.18160375952720642, | |
| "learning_rate": 3.7651019814126654e-05, | |
| "loss": 10.3409, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.28421052631578947, | |
| "eval_loss": 10.34239673614502, | |
| "eval_runtime": 0.3751, | |
| "eval_samples_per_second": 373.281, | |
| "eval_steps_per_second": 186.64, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 0.28547802567481995, | |
| "learning_rate": 2.8165064990227252e-05, | |
| "loss": 10.3431, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.3007518796992481, | |
| "grad_norm": 0.15805752575397491, | |
| "learning_rate": 1.985863781320435e-05, | |
| "loss": 10.3395, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3157894736842105, | |
| "grad_norm": 0.14654022455215454, | |
| "learning_rate": 1.286812958766106e-05, | |
| "loss": 10.3429, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.3308270676691729, | |
| "grad_norm": 0.13083398342132568, | |
| "learning_rate": 7.308324265397836e-06, | |
| "loss": 10.3441, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.3458646616541353, | |
| "grad_norm": 0.2036089301109314, | |
| "learning_rate": 3.270513696097055e-06, | |
| "loss": 10.3387, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.3609022556390977, | |
| "grad_norm": 0.13818128407001495, | |
| "learning_rate": 8.209986176753948e-07, | |
| "loss": 10.3463, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.37593984962406013, | |
| "grad_norm": 0.20554596185684204, | |
| "learning_rate": 0.0, | |
| "loss": 10.3422, | |
| "step": 250 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 250, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 63, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1634451456000.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |