| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0012675556456928458, | |
| "eval_steps": 3, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 5.070222582771384e-05, | |
| "grad_norm": 16.734760284423828, | |
| "learning_rate": 2e-05, | |
| "loss": 29.9648, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 5.070222582771384e-05, | |
| "eval_loss": 8.524748802185059, | |
| "eval_runtime": 1510.4845, | |
| "eval_samples_per_second": 5.498, | |
| "eval_steps_per_second": 2.749, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00010140445165542768, | |
| "grad_norm": 14.790090560913086, | |
| "learning_rate": 4e-05, | |
| "loss": 33.7103, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0001521066774831415, | |
| "grad_norm": 14.030793190002441, | |
| "learning_rate": 6e-05, | |
| "loss": 29.0882, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0001521066774831415, | |
| "eval_loss": 8.50710391998291, | |
| "eval_runtime": 1519.8925, | |
| "eval_samples_per_second": 5.464, | |
| "eval_steps_per_second": 2.732, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00020280890331085536, | |
| "grad_norm": 12.610204696655273, | |
| "learning_rate": 8e-05, | |
| "loss": 28.2266, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0002535111291385692, | |
| "grad_norm": 16.985925674438477, | |
| "learning_rate": 0.0001, | |
| "loss": 35.9331, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.000304213354966283, | |
| "grad_norm": 12.154075622558594, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 26.2052, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.000304213354966283, | |
| "eval_loss": 8.200469017028809, | |
| "eval_runtime": 1521.6515, | |
| "eval_samples_per_second": 5.458, | |
| "eval_steps_per_second": 2.729, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00035491558079399685, | |
| "grad_norm": 25.15923500061035, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 41.4569, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0004056178066217107, | |
| "grad_norm": 21.290430068969727, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 26.694, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0004563200324494245, | |
| "grad_norm": 23.049171447753906, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 29.4909, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0004563200324494245, | |
| "eval_loss": 6.978517055511475, | |
| "eval_runtime": 1521.0316, | |
| "eval_samples_per_second": 5.46, | |
| "eval_steps_per_second": 2.73, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0005070222582771384, | |
| "grad_norm": 26.38067054748535, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 34.6105, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0005577244841048522, | |
| "grad_norm": 24.686254501342773, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 23.0639, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.000608426709932566, | |
| "grad_norm": 22.281373977661133, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 21.833, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.000608426709932566, | |
| "eval_loss": 5.586967945098877, | |
| "eval_runtime": 1519.8132, | |
| "eval_samples_per_second": 5.464, | |
| "eval_steps_per_second": 2.733, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0006591289357602798, | |
| "grad_norm": 164.1787567138672, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 20.3158, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0007098311615879937, | |
| "grad_norm": 23.327159881591797, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 21.234, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0007605333874157076, | |
| "grad_norm": 17.704011917114258, | |
| "learning_rate": 5e-05, | |
| "loss": 13.4687, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0007605333874157076, | |
| "eval_loss": 4.6250200271606445, | |
| "eval_runtime": 1519.2183, | |
| "eval_samples_per_second": 5.467, | |
| "eval_steps_per_second": 2.734, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0008112356132434214, | |
| "grad_norm": 21.224292755126953, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 16.2957, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0008619378390711352, | |
| "grad_norm": 27.829790115356445, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 23.9581, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.000912640064898849, | |
| "grad_norm": 27.01038360595703, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 19.2738, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.000912640064898849, | |
| "eval_loss": 3.9614384174346924, | |
| "eval_runtime": 1519.4201, | |
| "eval_samples_per_second": 5.466, | |
| "eval_steps_per_second": 2.733, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0009633422907265629, | |
| "grad_norm": 31.614938735961914, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 16.2201, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0010140445165542768, | |
| "grad_norm": 30.133304595947266, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 13.2166, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0010647467423819906, | |
| "grad_norm": 30.905317306518555, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 17.1181, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0010647467423819906, | |
| "eval_loss": 3.6349897384643555, | |
| "eval_runtime": 1521.227, | |
| "eval_samples_per_second": 5.459, | |
| "eval_steps_per_second": 2.73, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0011154489682097045, | |
| "grad_norm": 30.500104904174805, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 11.5841, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0011661511940374183, | |
| "grad_norm": 33.284156799316406, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 17.699, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.001216853419865132, | |
| "grad_norm": 22.113712310791016, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 9.9624, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.001216853419865132, | |
| "eval_loss": 3.54598331451416, | |
| "eval_runtime": 1519.2854, | |
| "eval_samples_per_second": 5.466, | |
| "eval_steps_per_second": 2.734, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0012675556456928458, | |
| "grad_norm": 35.23492431640625, | |
| "learning_rate": 0.0, | |
| "loss": 14.2806, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 25, | |
| "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": 2.390484150386688e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |