| { | |
| "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": 6.0979489269369775, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0039, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.134863581637819, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.9949, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 5.292029104723269, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.8876, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.2205521190530266, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.8951, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.276234836711806, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.8828, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.985652394852927, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9167, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.545257523911323, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8374, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 3.663289883392731, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.8489, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 3.0625304779189917, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8144, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 2.378716706889852, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.7667, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 3.3474745135076995, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 1.2105, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.6691230927678793, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.8045, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.584681258007362, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.7048, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 1.3071235014688336, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.6977, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.0246032277537351, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.6766, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.2496403269529532, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.6365, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.9454030382856026, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.6892, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.901771162802523, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.642, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.8300716616839995, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.6058, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.7848959406401916, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.7035, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 1.3362509866316135, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 1.0084, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.7402413575120739, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.5898, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.6537655393817654, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.5657, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.9051309121804384, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.5759, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.588104979379416, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.5189, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.7756674073769759, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.5558, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.5773378739796524, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.4982, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.5889777243008133, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.5218, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.6459385681535833, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.5441, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.5429104695491352, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.4949, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.8308358406398794, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.6658, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.9408487887928983, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.6883, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.5560982348627105, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.4425, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.6089828844140907, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.4254, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.6053182053596003, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.4313, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.5619854518193421, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.4369, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.7009813930231176, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.4152, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.6367855159468767, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.4221, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.5155109492447385, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.4276, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.5952723378874096, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.3972, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.6467231049168265, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.4193, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.8471796671297391, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.6428, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.5722839030090756, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.3635, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.6048564774047649, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.3413, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.4710199214479688, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.3568, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.5398965637672368, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.3561, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.5902242932378647, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.3559, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.44488163004279835, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.3096, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.4780505937230667, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.3643, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.42016490749961916, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.3227, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.4260842130373242, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.3249, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.7337740451426735, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.4955, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.554816414282422, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.3366, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.48531123883764515, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.3044, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.4307665645670527, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.2914, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.46330056018315496, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.3168, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.5209097641633441, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.3132, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.4725637148843491, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.2811, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.4584258838700549, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.2883, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.4554633309437899, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.3026, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.37919046874502776, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.2825, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.3671404570095355, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.2674, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.702597048628452, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.4628, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.4202010393528316, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.3168, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.3887199099980158, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.2553, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.3687166113682504, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.2831, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.3800211212288531, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.2761, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.39914490181521206, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.2766, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.3572708895852964, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.2734, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.3603887669060873, | |
| "learning_rate": 0.0, | |
| "loss": 0.2677, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 1.1887870124385894e+17, | |
| "train_loss": 0.5270571793828692, | |
| "train_runtime": 5256.4785, | |
| "train_samples_per_second": 1.332, | |
| "train_steps_per_second": 0.013 | |
| } | |
| ], | |
| "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": 1.1887870124385894e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |