| { | |
| "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": 6.710640352747571, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0333, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.40275847985125, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0043, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 6.049959658237764, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0453, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.548113121337474, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.8896, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.407187427056272, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.8908, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 5.258123937609197, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.963, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 5.275942161463968, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9012, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 4.503607715951538, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.901, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 3.8640049095809954, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.833, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 3.1525904034742047, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8863, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 4.2496791693260185, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 1.2868, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 2.466127270071338, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.7664, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.9558217778358402, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.7523, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 1.7048429279785347, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7023, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.1109578288615494, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.7067, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.2348676027721337, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7817, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 1.106651929425531, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.7074, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.9316377701201336, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.668, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.873254032788255, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.6469, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.9594336329079994, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.7406, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 1.1898669166484566, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.9693, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 1.0713281320888448, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.6007, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.7618126965196705, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.5184, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.8111088444601813, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.6223, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.9550822968252537, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.5143, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.776593890441419, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.5261, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.9394992923993132, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.5024, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 1.2442163339205812, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.5693, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 5.993768681452137, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.5067, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 4.647669311739557, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.5676, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 245.09533752108203, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.7217, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 8.412961770127216, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.6955, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 1.537482121757925, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.4832, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.8993771031891318, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.4543, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.9640900579874829, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.4477, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.8088415215060272, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.3905, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.9856853006393917, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.4573, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.8927314165554789, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.4505, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.7975364777630238, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.3776, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.8010856564865362, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.3832, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.7243354896864755, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.467, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 1.1124589701633, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.584, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.7956261301192588, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.3959, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.647986031435044, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.3556, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.6438498076667752, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.3398, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.5976715353540674, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.2832, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.8230442762312475, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.3705, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.7161477745028793, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.3326, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.6162454357513472, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.32, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.7089915792860834, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.3498, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.6101541426889403, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.3345, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.8308848421643276, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.4379, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.7537236225602933, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.3199, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.5857897868002195, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.2781, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.5805011095791265, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.2717, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.5447460062635734, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.288, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.544809880445409, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.2583, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.6694962622808126, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.2956, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.596161232918502, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.2708, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.5434389788773799, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.2803, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.4932004334803732, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.2573, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.44850979356424797, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.2566, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.7481469091597499, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.3587, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.5232478737887413, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.2913, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.4819924683048751, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.2825, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.40867432271179827, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.2183, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.45393690953549853, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.2659, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.477401139871737, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.2609, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.43766396340045904, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.2306, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.4619792901049927, | |
| "learning_rate": 0.0, | |
| "loss": 0.2328, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 7.745076917908275e+16, | |
| "train_loss": 0.5336216585976737, | |
| "train_runtime": 3308.8816, | |
| "train_samples_per_second": 2.116, | |
| "train_steps_per_second": 0.021 | |
| } | |
| ], | |
| "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": 7.745076917908275e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |