| { | |
| "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": 6.138772311732421, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.8474, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.149121924772074, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.8063, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 5.7170060260680025, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.828, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.5687493691076018, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.7264, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 3.8642553056598046, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.7479, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.27913691485856, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.739, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.636569313872795, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6759, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 3.2590498783579314, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.6897, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.6698179828150144, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.6322, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.9797683413659029, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.5988, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 1.9197640044877222, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.9058, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.4470352970044318, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.5491, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 0.9973657454394809, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.612, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 0.8260156008229599, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.5557, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.7717343116645107, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.5059, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 0.7189351502924612, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.5427, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.7574687449482682, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.5032, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.6895022065240816, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.5721, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.555485283272524, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.4093, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.7752483457142532, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.5748, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 0.987204958223964, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.7393, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.6455566021373152, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.4516, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.656846131772044, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.4326, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.6329013965170377, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.4429, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.6117180471829058, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.4149, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.628907132841074, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.4562, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.5250754265105443, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.3801, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.5528856301359142, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.4528, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.49762062065008617, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.3652, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.5678150085295867, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.3941, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.7182653561737806, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.4997, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.7198580521928916, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.5256, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.55613964146649, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.3376, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.44654590379677334, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.3413, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.49470346392168607, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.3297, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.6359373762810614, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.3935, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.5120907067985542, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.3089, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.52280757480621, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.2972, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.5170294858794615, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.3712, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.5258770035759092, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.3465, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.5441957181295032, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.3384, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.8170462099370533, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.4622, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.5372319727338881, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.316, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.501423853993971, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.2768, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.582226515726494, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.3048, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.534873737230733, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.2831, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.44445567908069394, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.2499, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.48257564402922554, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.2893, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.41987161750119684, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.2642, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.4256869624833546, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.2448, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.4360390945249442, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.2981, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.540259076399103, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.3753, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.5187414315313741, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.27, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.4049034344700195, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.2394, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.40335521295429977, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.2231, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.42184289350130527, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.2481, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.39019489904715104, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.2452, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.35586513693082694, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.247, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.33848914106784656, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.2202, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.36631685899692323, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.2257, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.39835915269990335, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.2663, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.3617646269200946, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.2051, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.631661736988787, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.3921, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.32121955239320144, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.2074, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.3217555962213745, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.2196, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.30278358293136176, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.195, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.330341196912622, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.2468, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.331871478404907, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.2501, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.31888309956853617, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.2248, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.3338433943049503, | |
| "learning_rate": 0.0, | |
| "loss": 0.22, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 1.0793027218335334e+17, | |
| "train_loss": 0.419307304918766, | |
| "train_runtime": 5019.0542, | |
| "train_samples_per_second": 1.395, | |
| "train_steps_per_second": 0.014 | |
| } | |
| ], | |
| "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": 1.0793027218335334e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |