| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 30.0, | |
| "eval_steps": 500, | |
| "global_step": 60, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.40977156162261963, | |
| "learning_rate": 4e-05, | |
| "loss": 0.6411, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.42220255732536316, | |
| "learning_rate": 8e-05, | |
| "loss": 0.6556, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.29474809765815735, | |
| "learning_rate": 0.00012, | |
| "loss": 0.6299, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.22392094135284424, | |
| "learning_rate": 0.00016, | |
| "loss": 0.6031, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.2246803194284439, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5672, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.1952209174633026, | |
| "learning_rate": 0.00019636363636363636, | |
| "loss": 0.531, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "grad_norm": 0.2026151567697525, | |
| "learning_rate": 0.00019272727272727274, | |
| "loss": 0.4935, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.20893992483615875, | |
| "learning_rate": 0.0001890909090909091, | |
| "loss": 0.4368, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "grad_norm": 0.21967507898807526, | |
| "learning_rate": 0.00018545454545454545, | |
| "loss": 0.3984, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.2647532820701599, | |
| "learning_rate": 0.00018181818181818183, | |
| "loss": 0.3768, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "grad_norm": 0.24694104492664337, | |
| "learning_rate": 0.0001781818181818182, | |
| "loss": 0.3285, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.3180811405181885, | |
| "learning_rate": 0.00017454545454545454, | |
| "loss": 0.2968, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "grad_norm": 0.2574905753135681, | |
| "learning_rate": 0.0001709090909090909, | |
| "loss": 0.2625, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.5857559442520142, | |
| "learning_rate": 0.00016727272727272728, | |
| "loss": 0.2228, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 0.24397815763950348, | |
| "learning_rate": 0.00016363636363636366, | |
| "loss": 0.2072, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.2656276524066925, | |
| "learning_rate": 0.00016, | |
| "loss": 0.1601, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "grad_norm": 0.24156434834003448, | |
| "learning_rate": 0.00015636363636363637, | |
| "loss": 0.1538, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.21211856603622437, | |
| "learning_rate": 0.00015272727272727275, | |
| "loss": 0.139, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "grad_norm": 0.22596681118011475, | |
| "learning_rate": 0.0001490909090909091, | |
| "loss": 0.1297, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.3587157428264618, | |
| "learning_rate": 0.00014545454545454546, | |
| "loss": 0.1147, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 10.5, | |
| "grad_norm": 0.11396785080432892, | |
| "learning_rate": 0.00014181818181818184, | |
| "loss": 0.0921, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.16308319568634033, | |
| "learning_rate": 0.0001381818181818182, | |
| "loss": 0.1176, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 11.5, | |
| "grad_norm": 0.10265325754880905, | |
| "learning_rate": 0.00013454545454545455, | |
| "loss": 0.0933, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.12172083556652069, | |
| "learning_rate": 0.00013090909090909093, | |
| "loss": 0.0948, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "grad_norm": 0.1143866702914238, | |
| "learning_rate": 0.00012727272727272728, | |
| "loss": 0.088, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.10438387840986252, | |
| "learning_rate": 0.00012363636363636364, | |
| "loss": 0.0815, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "grad_norm": 0.09685764461755753, | |
| "learning_rate": 0.00012, | |
| "loss": 0.0796, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.09470539540052414, | |
| "learning_rate": 0.00011636363636363636, | |
| "loss": 0.0731, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 14.5, | |
| "grad_norm": 0.08961515873670578, | |
| "learning_rate": 0.00011272727272727272, | |
| "loss": 0.0635, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.09088480472564697, | |
| "learning_rate": 0.00010909090909090909, | |
| "loss": 0.0735, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 15.5, | |
| "grad_norm": 0.08304226398468018, | |
| "learning_rate": 0.00010545454545454545, | |
| "loss": 0.0608, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.08834656327962875, | |
| "learning_rate": 0.00010181818181818181, | |
| "loss": 0.0635, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 16.5, | |
| "grad_norm": 0.0853792354464531, | |
| "learning_rate": 9.818181818181818e-05, | |
| "loss": 0.0556, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.08675861358642578, | |
| "learning_rate": 9.454545454545455e-05, | |
| "loss": 0.0579, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 17.5, | |
| "grad_norm": 0.09384719282388687, | |
| "learning_rate": 9.090909090909092e-05, | |
| "loss": 0.0551, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 0.08688965439796448, | |
| "learning_rate": 8.727272727272727e-05, | |
| "loss": 0.0481, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 18.5, | |
| "grad_norm": 0.08417191356420517, | |
| "learning_rate": 8.363636363636364e-05, | |
| "loss": 0.0461, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.0866803452372551, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0478, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 19.5, | |
| "grad_norm": 0.08191067725419998, | |
| "learning_rate": 7.636363636363637e-05, | |
| "loss": 0.0451, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.08162440359592438, | |
| "learning_rate": 7.272727272727273e-05, | |
| "loss": 0.0401, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 20.5, | |
| "grad_norm": 0.0843905583024025, | |
| "learning_rate": 6.90909090909091e-05, | |
| "loss": 0.0343, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "grad_norm": 0.09161655604839325, | |
| "learning_rate": 6.545454545454546e-05, | |
| "loss": 0.0429, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 21.5, | |
| "grad_norm": 0.08548980951309204, | |
| "learning_rate": 6.181818181818182e-05, | |
| "loss": 0.0364, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "grad_norm": 0.09480360895395279, | |
| "learning_rate": 5.818181818181818e-05, | |
| "loss": 0.0336, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 22.5, | |
| "grad_norm": 0.08082766085863113, | |
| "learning_rate": 5.4545454545454546e-05, | |
| "loss": 0.0313, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "grad_norm": 0.08921442180871964, | |
| "learning_rate": 5.090909090909091e-05, | |
| "loss": 0.0321, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 23.5, | |
| "grad_norm": 0.08190133422613144, | |
| "learning_rate": 4.7272727272727275e-05, | |
| "loss": 0.0289, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 0.08818236738443375, | |
| "learning_rate": 4.3636363636363636e-05, | |
| "loss": 0.0287, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 24.5, | |
| "grad_norm": 0.0787026658654213, | |
| "learning_rate": 4e-05, | |
| "loss": 0.027, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 0.08572469651699066, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.0252, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 25.5, | |
| "grad_norm": 0.0922461524605751, | |
| "learning_rate": 3.272727272727273e-05, | |
| "loss": 0.0265, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "grad_norm": 0.09386759996414185, | |
| "learning_rate": 2.909090909090909e-05, | |
| "loss": 0.0218, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 26.5, | |
| "grad_norm": 0.09283447265625, | |
| "learning_rate": 2.5454545454545454e-05, | |
| "loss": 0.0231, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "grad_norm": 0.08825419098138809, | |
| "learning_rate": 2.1818181818181818e-05, | |
| "loss": 0.0212, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 27.5, | |
| "grad_norm": 0.08191809058189392, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 0.0217, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "grad_norm": 0.08192411810159683, | |
| "learning_rate": 1.4545454545454545e-05, | |
| "loss": 0.0196, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 28.5, | |
| "grad_norm": 0.09538112580776215, | |
| "learning_rate": 1.0909090909090909e-05, | |
| "loss": 0.0202, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "grad_norm": 0.09735918790102005, | |
| "learning_rate": 7.272727272727272e-06, | |
| "loss": 0.019, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 29.5, | |
| "grad_norm": 0.08030348271131516, | |
| "learning_rate": 3.636363636363636e-06, | |
| "loss": 0.0171, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 0.0887613520026207, | |
| "learning_rate": 0.0, | |
| "loss": 0.0207, | |
| "step": 60 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 60, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7072322089623552.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |