| { | |
| "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": 6.311398934406293, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0243, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.360546518893127, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0262, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 5.930357722083423, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.032, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.798628679340223, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9695, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 3.983189535982094, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9351, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.045173870148383, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9606, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 3.8612792578536324, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8899, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 2.838067047899917, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.8878, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.230211996796827, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8421, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.574696486999752, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8129, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 1.9212582457649323, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 1.2468, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.2501622088262347, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.7543, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.2603432806950696, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.8058, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 0.9553667896367697, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7128, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.9775412909139116, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.7221, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 0.9739905673082503, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7231, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.8782335809604456, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.72, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.7759555858893842, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.7197, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.8138713876868261, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.7162, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.9059553759410488, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.7296, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 1.237690610809914, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 1.0598, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.7205027334039861, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.6682, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.8720921789399912, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.6209, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.696896822200175, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.6522, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.8863861364518416, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6545, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.7149708041381727, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.6445, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.7123723465839348, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.5958, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.5546693237538727, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.6468, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.6976563112936924, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.6573, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.5455835226792909, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.5916, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.5754472650949384, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.7711, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.7427574588874565, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.8452, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.5189460534085905, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.5377, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.4687437000446221, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.6047, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.5832287424726785, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.5977, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.47974519577367125, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.5517, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.5977516379403678, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.5795, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.44940309997432093, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.5281, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.4497212107115964, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.5564, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.45725005737647556, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.558, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.4565870761371436, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.5639, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.6649655822862885, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.8799, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.5047863010629273, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.5312, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.5125404615568481, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.5356, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.45671807214139576, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.5193, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.5049825311016052, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.5151, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.38408061949764255, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.5254, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.4832738609152666, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.5147, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.4359758304002188, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.5156, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.3203646835327328, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.4949, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.4532770248158105, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.5145, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.72916169586045, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.7746, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.4277648127261752, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.5344, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.37151643922260846, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.4932, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.3467460865194236, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.4801, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.37179830017184623, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.4883, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.328751089946516, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.4914, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.30532166927152, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.4807, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.3269699835342485, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.5066, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.3159994132616735, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.494, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.3098446905283662, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.4997, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.3053835003373171, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.4909, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.5008981360282949, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.7545, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.28706103534364785, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.485, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.2680979714347468, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.4693, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.2757258320827127, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.4895, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.2809461124466278, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.4795, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.2547733341811787, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.4899, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.2545556694744232, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.475, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.2623738010476276, | |
| "learning_rate": 0.0, | |
| "loss": 0.4752, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 2.3192243876449485e+17, | |
| "train_loss": 0.6587781152554921, | |
| "train_runtime": 8562.2854, | |
| "train_samples_per_second": 0.818, | |
| "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": 2.3192243876449485e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |