| { | |
| "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": 6.290681887421301, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.8197, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.4387706801119595, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.8262, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 5.610593037439924, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.7858, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.496544343753618, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.7474, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 3.7657151253378, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.7278, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.141692581035384, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.7151, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.291807151738781, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6884, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 2.414585459738222, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.6609, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.477525179599756, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.6633, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.8158864190707462, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.6068, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 2.2739283315902865, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.9551, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.1888515673635605, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.5914, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.0653438165059894, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.5584, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 0.8563033027884882, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.5462, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.9348261956041793, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.5634, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 0.8736293685809896, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.5486, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.726938638385407, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.5487, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.7524443174181714, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.5331, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.7201348102664357, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.5384, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.7068056490161952, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.5059, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 0.8056751148007627, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.8264, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.5190824490104428, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.4938, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.6199432042932124, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.4771, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.5184114358073176, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.4933, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.5095996300615131, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.4711, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.5590012636215204, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.4845, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.43706053155279967, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.4578, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.5418894185197094, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.483, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.4173470381096769, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.4474, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.4495865695732863, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.4472, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.54478000427382, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.5771, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.5467467983622764, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.59, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.41747936887094866, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.4507, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.4559991736895225, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.437, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.3811813903464245, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.4173, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.43627508968414663, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.432, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.40699718953150793, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.4153, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.40181254181994597, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.4144, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.4200141356854101, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.4197, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.36062698717541253, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.4065, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.3728967039649329, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.398, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.5686369159459169, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.6421, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.3670347086034741, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.3989, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.39705745036823337, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.3902, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.3330464541643674, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.3804, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.37775173331151735, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.3957, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.3050464850842139, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.3907, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.3122694003201031, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.3703, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.32200310979769975, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.3866, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.744976469973927, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.3885, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.2723333105526255, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.3582, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.4478149188224321, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.565, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.30159763111773946, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.4244, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.2640933900639159, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.375, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.24626512665269715, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.363, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.256618236979838, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.3525, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.2660584468282525, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.3575, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.23708254687585162, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.3656, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.2301165661348355, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.3777, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.2199543544985608, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.3568, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.2326012545282132, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.3759, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.20964519570983292, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.3414, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.3925735610093721, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.5773, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.21739967518872216, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.3686, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.1946591003891265, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.3469, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.2029807462480628, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.342, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.20595913602883442, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.3542, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.21317197245584113, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.3656, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.21063570478932825, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.3687, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.20153975238827632, | |
| "learning_rate": 0.0, | |
| "loss": 0.3518, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 2.737828979183452e+17, | |
| "train_loss": 0.4943068653345108, | |
| "train_runtime": 8357.6844, | |
| "train_samples_per_second": 0.838, | |
| "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": 2.737828979183452e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |