| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 27.0, | |
| "eval_steps": 500, | |
| "global_step": 270, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5333333333333333, | |
| "grad_norm": 0.3464358448982239, | |
| "learning_rate": 7.142857142857143e-05, | |
| "loss": 0.7933, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.2775850296020508, | |
| "learning_rate": 0.00014285714285714287, | |
| "loss": 0.6763, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.5333333333333332, | |
| "grad_norm": 0.20196351408958435, | |
| "learning_rate": 0.00019999247018391447, | |
| "loss": 0.612, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.20944692194461823, | |
| "learning_rate": 0.00019972904566786903, | |
| "loss": 0.593, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.533333333333333, | |
| "grad_norm": 0.1796199381351471, | |
| "learning_rate": 0.000199090263542778, | |
| "loss": 0.528, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.3253108859062195, | |
| "learning_rate": 0.00019807852804032305, | |
| "loss": 0.4901, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.533333333333333, | |
| "grad_norm": 0.2155134528875351, | |
| "learning_rate": 0.00019669764710448522, | |
| "loss": 0.4385, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.40249642729759216, | |
| "learning_rate": 0.00019495281805930367, | |
| "loss": 0.4146, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.533333333333333, | |
| "grad_norm": 0.21931566298007965, | |
| "learning_rate": 0.00019285060804732158, | |
| "loss": 0.3422, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.47045040130615234, | |
| "learning_rate": 0.00019039892931234435, | |
| "loss": 0.334, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 5.533333333333333, | |
| "grad_norm": 0.33896583318710327, | |
| "learning_rate": 0.00018760700941954065, | |
| "loss": 0.3055, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.6106130480766296, | |
| "learning_rate": 0.00018448535652497073, | |
| "loss": 0.2364, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 6.533333333333333, | |
| "grad_norm": 0.34181615710258484, | |
| "learning_rate": 0.0001810457198252595, | |
| "loss": 0.19, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.8446937799453735, | |
| "learning_rate": 0.0001773010453362737, | |
| "loss": 0.1965, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 7.533333333333333, | |
| "grad_norm": 0.5083507299423218, | |
| "learning_rate": 0.00017326542716724128, | |
| "loss": 0.1207, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.7833607196807861, | |
| "learning_rate": 0.0001689540544737067, | |
| "loss": 0.1283, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 8.533333333333333, | |
| "grad_norm": 0.3553655743598938, | |
| "learning_rate": 0.00016438315428897915, | |
| "loss": 0.0811, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.7091866135597229, | |
| "learning_rate": 0.00015956993044924334, | |
| "loss": 0.0802, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 9.533333333333333, | |
| "grad_norm": 0.43794405460357666, | |
| "learning_rate": 0.00015453249884220464, | |
| "loss": 0.0492, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.5798842310905457, | |
| "learning_rate": 0.00014928981922297842, | |
| "loss": 0.0524, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 10.533333333333333, | |
| "grad_norm": 0.3912927210330963, | |
| "learning_rate": 0.00014386162385385278, | |
| "loss": 0.0302, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.4159412682056427, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 0.0328, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 11.533333333333333, | |
| "grad_norm": 0.24435237050056458, | |
| "learning_rate": 0.0001325310292162263, | |
| "loss": 0.0239, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.24223262071609497, | |
| "learning_rate": 0.00012667127574748986, | |
| "loss": 0.0171, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 12.533333333333333, | |
| "grad_norm": 0.1762484759092331, | |
| "learning_rate": 0.00012071113761922186, | |
| "loss": 0.0135, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.28046339750289917, | |
| "learning_rate": 0.00011467304744553618, | |
| "loss": 0.0143, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 13.533333333333333, | |
| "grad_norm": 0.28064408898353577, | |
| "learning_rate": 0.000108579731234444, | |
| "loss": 0.0128, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.19103752076625824, | |
| "learning_rate": 0.00010245412285229124, | |
| "loss": 0.0101, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 14.533333333333333, | |
| "grad_norm": 0.2761705815792084, | |
| "learning_rate": 9.631927770586412e-05, | |
| "loss": 0.0096, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.3451383709907532, | |
| "learning_rate": 9.019828596704394e-05, | |
| "loss": 0.0084, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 15.533333333333333, | |
| "grad_norm": 0.22912859916687012, | |
| "learning_rate": 8.411418566661388e-05, | |
| "loss": 0.0087, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.2961307466030121, | |
| "learning_rate": 7.808987598431303e-05, | |
| "loss": 0.0054, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 16.533333333333335, | |
| "grad_norm": 0.05174371972680092, | |
| "learning_rate": 7.21480310614947e-05, | |
| "loss": 0.0037, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.6668692827224731, | |
| "learning_rate": 6.6311014660778e-05, | |
| "loss": 0.0086, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 17.533333333333335, | |
| "grad_norm": 0.0423920676112175, | |
| "learning_rate": 6.060079599389521e-05, | |
| "loss": 0.0046, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 0.048323437571525574, | |
| "learning_rate": 5.503886703453933e-05, | |
| "loss": 0.005, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 18.533333333333335, | |
| "grad_norm": 0.14866454899311066, | |
| "learning_rate": 4.964616162742826e-05, | |
| "loss": 0.0039, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.04198712110519409, | |
| "learning_rate": 4.444297669803981e-05, | |
| "loss": 0.0033, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 19.533333333333335, | |
| "grad_norm": 0.037916265428066254, | |
| "learning_rate": 3.944889585956746e-05, | |
| "loss": 0.0075, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.031426794826984406, | |
| "learning_rate": 3.468271570462235e-05, | |
| "loss": 0.0036, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 20.533333333333335, | |
| "grad_norm": 0.1819785088300705, | |
| "learning_rate": 3.016237505910272e-05, | |
| "loss": 0.0062, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "grad_norm": 0.09462948143482208, | |
| "learning_rate": 2.5904887464504114e-05, | |
| "loss": 0.0034, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 21.533333333333335, | |
| "grad_norm": 0.11672957241535187, | |
| "learning_rate": 2.1926277142790552e-05, | |
| "loss": 0.0025, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "grad_norm": 0.04571106657385826, | |
| "learning_rate": 1.824151868484164e-05, | |
| "loss": 0.0042, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 22.533333333333335, | |
| "grad_norm": 0.019737839698791504, | |
| "learning_rate": 1.486448068947348e-05, | |
| "loss": 0.0021, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "grad_norm": 0.03002820909023285, | |
| "learning_rate": 1.1807873565164506e-05, | |
| "loss": 0.0049, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 23.533333333333335, | |
| "grad_norm": 0.24355602264404297, | |
| "learning_rate": 9.083201690947763e-06, | |
| "loss": 0.004, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 0.035868410021066666, | |
| "learning_rate": 6.700720116526116e-06, | |
| "loss": 0.0024, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 24.533333333333335, | |
| "grad_norm": 0.029756512492895126, | |
| "learning_rate": 4.669395964580614e-06, | |
| "loss": 0.0025, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 0.2836012542247772, | |
| "learning_rate": 2.996874680545603e-06, | |
| "loss": 0.0059, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 25.533333333333335, | |
| "grad_norm": 0.2209901213645935, | |
| "learning_rate": 1.6894512568783716e-06, | |
| "loss": 0.004, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "grad_norm": 0.023770242929458618, | |
| "learning_rate": 7.520465401290033e-07, | |
| "loss": 0.0033, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 26.533333333333335, | |
| "grad_norm": 0.02641889452934265, | |
| "learning_rate": 1.8818870998508208e-07, | |
| "loss": 0.0032, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "grad_norm": 0.03176680952310562, | |
| "learning_rate": 0.0, | |
| "loss": 0.0025, | |
| "step": 270 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 270, | |
| "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": 1.3785203655910195e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |