| { | |
| "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": 6.623335257556795, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0754, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.468666716198513, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0398, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 5.808781333127986, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0212, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.4009438995676167, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9451, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.293229714535669, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9198, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 5.048882803420056, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9272, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 5.3161041074067485, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9403, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 4.142711064105125, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.958, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 3.5552409143014807, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8422, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 2.914900926336261, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8502, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 3.9776945706457054, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 1.3371, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.687162679708264, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.748, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.9050625901098388, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.8296, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 1.207750787190213, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.6878, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.0281363664491794, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.7225, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.3527797297903204, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7726, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.7759867509359699, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.62, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.7997572536131939, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.6897, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.9339821538449918, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.7043, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.7556973442553836, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.7121, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 1.3268468894217489, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 1.024, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.8593808228449634, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.5686, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.9884332686111288, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.6147, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.9569862095049257, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.5593, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.9158829556487439, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.54, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 1.0555809054613994, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.5651, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.7650730663974188, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.5229, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.8069510012079995, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.5307, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.872184452484184, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.5363, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.707961216145777, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.5516, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.8638940885528759, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.5865, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 1.0144541069720305, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.637, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.6549218776230971, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.4121, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.6141727434064369, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.4298, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.7369664876395177, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.4259, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.6208544492230988, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.3611, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.6103010536778515, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.3966, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.7351074105276599, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.4578, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.5832300178328728, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.3728, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.6134036736050263, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.3919, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.5594049170410939, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.3509, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 1.012463749532614, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.5713, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.6623377493902274, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.3405, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.5317921925754261, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.2576, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.651689076675409, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.3201, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.7617989648073942, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.3028, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.6342739259273993, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.3064, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.511185806987183, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.2719, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.5509491730974162, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.2883, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.566879369388937, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.2763, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.56684332445241, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.3097, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.7801126268667066, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.38, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.7420626322777163, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.2907, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.4987032113631743, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.2435, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.493315463691155, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.2502, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.5306509949402234, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.2243, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.5761934452362507, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.2525, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.4941620257861265, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.2139, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.5408606798210193, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.2503, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.488916701769066, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.2347, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.44465668989246515, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.2041, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.43741629224137873, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.227, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.7797898801366672, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.3229, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.49218594018634104, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.244, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.47398234310342474, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.2274, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.4317294660583358, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.2096, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.4519213273896479, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.2103, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.4496003298685158, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.2109, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.44300375039103923, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.2007, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.44263621535615455, | |
| "learning_rate": 0.0, | |
| "loss": 0.2069, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 7.816496427512627e+16, | |
| "train_loss": 0.5146719762257167, | |
| "train_runtime": 3586.4619, | |
| "train_samples_per_second": 1.952, | |
| "train_steps_per_second": 0.02 | |
| } | |
| ], | |
| "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": 7.816496427512627e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |