| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.72, | |
| "eval_steps": 500, | |
| "global_step": 70, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.096, | |
| "grad_norm": 6.36753858086959, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0175, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.420390237548739, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0279, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 5.989010781075463, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0366, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.786320615270596, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9572, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 3.969242278364763, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.935, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.014980046500334, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9366, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 3.97270297865598, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9147, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 2.917184698669727, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.8711, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.136458986883208, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8576, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.6173410179522831, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8208, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 1.1847229212319594, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.7775, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.308694421855294, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.7799, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.0015085339600847, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.746, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 1.2833602097502832, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7632, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.1282255134188002, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.744, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.2207618764992945, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7304, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.7990055999798287, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.6995, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.9963774599934593, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.7077, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.8835648576970384, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.7063, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.9539479663018313, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.709, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 0.7370787800507452, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.6841, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.788339597847025, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.65, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.7030903000095328, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.658, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.5988832049548954, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.6298, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.7139795353289241, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6483, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.692391447630256, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.6309, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.5739665724348132, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.6434, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.5610810198252421, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.6293, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.4952636831142249, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.6267, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.5280107338442572, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.6336, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.495982625659485, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.6135, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.5167479482075119, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.5747, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.49024935013630344, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.5785, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.420850869657489, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.565, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.47329944436501564, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.5754, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.4308276130004763, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.5689, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.4944106198353437, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.5739, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.44726454198687016, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.5675, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.5805033163878545, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.5609, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.4907858228959922, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.5667, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.4294389877778075, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.5535, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.5278391653597279, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.5497, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.4886296168056841, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.5144, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.4079493263552464, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.5346, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.424255679981601, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.5089, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.46030980397914745, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.5212, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.40123909296276317, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.5023, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.40965915379600076, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.524, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.3350337027588883, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.5127, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.39237590044095216, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.5173, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.42088876727236524, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.5175, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.3422825380016223, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.5196, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.3103201794741115, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.5167, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.36565293363509505, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.4949, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.3236845465444688, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.4852, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.3132290917910237, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.4737, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.31571415016443266, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.4791, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.30770618525793547, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.4832, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.3087437934349157, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.4959, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.2910723807268125, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.5013, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.303509594722706, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.491, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.2697484482457612, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.4854, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.29023076698085615, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.4771, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.27517776723600307, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.4942, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.2737934049377067, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.4731, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.26000595197849785, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.4638, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.2484002916999026, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.4797, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.25996061788621616, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.4891, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.24577235509430695, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.4878, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.26853650741851653, | |
| "learning_rate": 0.0, | |
| "loss": 0.4775, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 2.6888364113893786e+17, | |
| "train_loss": 0.6277445639882769, | |
| "train_runtime": 8990.8317, | |
| "train_samples_per_second": 0.779, | |
| "train_steps_per_second": 0.008 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 70, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "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": 2.6888364113893786e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |