| { | |
| "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": 6.759870972997541, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0561, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.884617683001761, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0619, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 6.294303475274697, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0336, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 3.026005939406625, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9729, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.188171866204751, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9467, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.278846700857285, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9552, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.363059281084836, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9082, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 2.962598755600655, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.8689, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.682728475363585, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8601, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.8284758448026284, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.821, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 2.452195045513616, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 1.2616, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.9555358343924083, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.814, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.3930043585076977, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.6836, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 1.3970047080242913, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7679, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.4110918890771027, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.7134, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.3846929125001388, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7895, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 1.6014778565857026, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.6977, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 1.035231227869965, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.6817, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 1.2947914714947253, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.6951, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.9127858779222362, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.7267, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 1.6985109878698759, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 1.1335, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.8073510399340955, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.6487, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.8561675891568324, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.6617, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.9657931422840813, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.6684, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.7405344620180114, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6159, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.8438611437209967, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.6346, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.7837151012762897, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.6347, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.7263114887844538, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.629, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.6774938508060868, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.6291, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.6331655899043057, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.634, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.6804468315641492, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.7152, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.887142112435411, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.8133, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.6527770304816765, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.5818, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.6612339388277001, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.5716, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.5333201905863904, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.5437, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.6409980700070198, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.5801, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.6200622465983276, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.5514, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.5153462864268997, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.5519, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.47891632225871833, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.5381, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.48883462969928476, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.5597, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.490052867185594, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.5581, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.7413232706693859, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.8638, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.4529764717525765, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.533, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.40306120220963876, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.4547, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.4677284160827223, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.5336, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.45405773774808783, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.5178, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.4550370171319642, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.5057, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.5007110356419949, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.5094, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.4179178645291421, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.463, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.41775068445121855, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.5249, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.4381060611981884, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.4775, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.7676022545149046, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.7366, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.46291803257228525, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.5247, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.3882304658961749, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.4931, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.41266211301030475, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.4678, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.3947677523117267, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.4738, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.36263644372145226, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.4612, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.4078453548946576, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.4629, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.36444110226572407, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.4812, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.35615840440056074, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.4427, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.35211436145781716, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.4883, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.3308546537202275, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.4554, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.5327715243748539, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.7528, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.3150186728989669, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.4637, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.30549521832138393, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.4474, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.3204089348519548, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.4676, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.29073880389358775, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.4568, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.302725569633108, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.4936, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.29438561373604466, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.4327, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.31316708469434273, | |
| "learning_rate": 0.0, | |
| "loss": 0.4531, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 1.8234082071884595e+17, | |
| "train_loss": 0.651562267116138, | |
| "train_runtime": 7893.9925, | |
| "train_samples_per_second": 0.887, | |
| "train_steps_per_second": 0.009 | |
| } | |
| ], | |
| "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": 1.8234082071884595e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |