| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 3, | |
| "global_step": 33, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09090909090909091, | |
| "grad_norm": 0.13040611147880554, | |
| "learning_rate": 1e-05, | |
| "loss": 3.3054, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.09090909090909091, | |
| "eval_loss": 1.6250120401382446, | |
| "eval_runtime": 5.3483, | |
| "eval_samples_per_second": 1.87, | |
| "eval_steps_per_second": 0.374, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 0.1381409764289856, | |
| "learning_rate": 2e-05, | |
| "loss": 3.2584, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.2727272727272727, | |
| "grad_norm": 0.13030081987380981, | |
| "learning_rate": 3e-05, | |
| "loss": 3.2721, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.2727272727272727, | |
| "eval_loss": 1.6240813732147217, | |
| "eval_runtime": 5.3485, | |
| "eval_samples_per_second": 1.87, | |
| "eval_steps_per_second": 0.374, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 0.14820316433906555, | |
| "learning_rate": 4e-05, | |
| "loss": 3.2555, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.45454545454545453, | |
| "grad_norm": 0.15054553747177124, | |
| "learning_rate": 5e-05, | |
| "loss": 3.3107, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.5454545454545454, | |
| "grad_norm": 0.1470218300819397, | |
| "learning_rate": 6e-05, | |
| "loss": 3.2923, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.5454545454545454, | |
| "eval_loss": 1.6198304891586304, | |
| "eval_runtime": 5.3572, | |
| "eval_samples_per_second": 1.867, | |
| "eval_steps_per_second": 0.373, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.6363636363636364, | |
| "grad_norm": 0.15020014345645905, | |
| "learning_rate": 7e-05, | |
| "loss": 3.1834, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.1556359976530075, | |
| "learning_rate": 8e-05, | |
| "loss": 3.223, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.8181818181818182, | |
| "grad_norm": 0.1711636781692505, | |
| "learning_rate": 9e-05, | |
| "loss": 3.1513, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.8181818181818182, | |
| "eval_loss": 1.6020313501358032, | |
| "eval_runtime": 5.3568, | |
| "eval_samples_per_second": 1.867, | |
| "eval_steps_per_second": 0.373, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 0.23649035394191742, | |
| "learning_rate": 0.0001, | |
| "loss": 3.2135, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.23038285970687866, | |
| "learning_rate": 9.953429730181653e-05, | |
| "loss": 3.138, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 0.2272975891828537, | |
| "learning_rate": 9.814586436738998e-05, | |
| "loss": 3.1555, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "eval_loss": 1.5541090965270996, | |
| "eval_runtime": 5.3607, | |
| "eval_samples_per_second": 1.865, | |
| "eval_steps_per_second": 0.373, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.1818181818181819, | |
| "grad_norm": 0.24618901312351227, | |
| "learning_rate": 9.586056507527266e-05, | |
| "loss": 3.0046, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.2727272727272727, | |
| "grad_norm": 0.23367342352867126, | |
| "learning_rate": 9.272097022732443e-05, | |
| "loss": 3.1042, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.3636363636363638, | |
| "grad_norm": 0.20990729331970215, | |
| "learning_rate": 8.8785564535221e-05, | |
| "loss": 3.0543, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.3636363636363638, | |
| "eval_loss": 1.5030436515808105, | |
| "eval_runtime": 5.3534, | |
| "eval_samples_per_second": 1.868, | |
| "eval_steps_per_second": 0.374, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 0.29300206899642944, | |
| "learning_rate": 8.412765716093272e-05, | |
| "loss": 3.0298, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.5454545454545454, | |
| "grad_norm": 0.2927170991897583, | |
| "learning_rate": 7.883401610574336e-05, | |
| "loss": 3.001, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.6363636363636362, | |
| "grad_norm": 0.29279839992523193, | |
| "learning_rate": 7.300325188655761e-05, | |
| "loss": 3.0495, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.6363636363636362, | |
| "eval_loss": 1.449854850769043, | |
| "eval_runtime": 5.3549, | |
| "eval_samples_per_second": 1.867, | |
| "eval_steps_per_second": 0.373, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.7272727272727273, | |
| "grad_norm": 0.27877599000930786, | |
| "learning_rate": 6.674398060854931e-05, | |
| "loss": 2.8904, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 0.3109539747238159, | |
| "learning_rate": 6.01728006526317e-05, | |
| "loss": 2.9107, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.9090909090909092, | |
| "grad_norm": 0.2679404020309448, | |
| "learning_rate": 5.341212066823355e-05, | |
| "loss": 2.9001, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.9090909090909092, | |
| "eval_loss": 1.3991481065750122, | |
| "eval_runtime": 5.366, | |
| "eval_samples_per_second": 1.864, | |
| "eval_steps_per_second": 0.373, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.2852272093296051, | |
| "learning_rate": 4.658787933176646e-05, | |
| "loss": 2.7902, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.090909090909091, | |
| "grad_norm": 0.3013867437839508, | |
| "learning_rate": 3.982719934736832e-05, | |
| "loss": 2.8728, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 0.3320194184780121, | |
| "learning_rate": 3.325601939145069e-05, | |
| "loss": 2.7332, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "eval_loss": 1.3631479740142822, | |
| "eval_runtime": 5.355, | |
| "eval_samples_per_second": 1.867, | |
| "eval_steps_per_second": 0.373, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.2727272727272725, | |
| "grad_norm": 0.3708702623844147, | |
| "learning_rate": 2.6996748113442394e-05, | |
| "loss": 2.7049, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.3636363636363638, | |
| "grad_norm": 0.3256295323371887, | |
| "learning_rate": 2.1165983894256647e-05, | |
| "loss": 2.9422, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.4545454545454546, | |
| "grad_norm": 0.3431509733200073, | |
| "learning_rate": 1.5872342839067306e-05, | |
| "loss": 2.6983, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.4545454545454546, | |
| "eval_loss": 1.3413066864013672, | |
| "eval_runtime": 5.3602, | |
| "eval_samples_per_second": 1.866, | |
| "eval_steps_per_second": 0.373, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 0.34741801023483276, | |
| "learning_rate": 1.1214435464779006e-05, | |
| "loss": 2.7043, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.6363636363636362, | |
| "grad_norm": 0.370277464389801, | |
| "learning_rate": 7.2790297726755716e-06, | |
| "loss": 2.5969, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 0.39455196261405945, | |
| "learning_rate": 4.139434924727359e-06, | |
| "loss": 2.7283, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.7272727272727275, | |
| "eval_loss": 1.3317468166351318, | |
| "eval_runtime": 5.3596, | |
| "eval_samples_per_second": 1.866, | |
| "eval_steps_per_second": 0.373, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.8181818181818183, | |
| "grad_norm": 0.4511341452598572, | |
| "learning_rate": 1.8541356326100433e-06, | |
| "loss": 2.5703, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": 0.3779066801071167, | |
| "learning_rate": 4.6570269818346224e-07, | |
| "loss": 2.6689, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.39674925804138184, | |
| "learning_rate": 0.0, | |
| "loss": 2.6383, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.3296959400177002, | |
| "eval_runtime": 5.3524, | |
| "eval_samples_per_second": 1.868, | |
| "eval_steps_per_second": 0.374, | |
| "step": 33 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 33, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.379135550291968e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |