| { | |
| "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": 7.633728896722896, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.2101, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 7.672393554899526, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.2277, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 6.781415412236372, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.1559, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.851136568272014, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 1.0984, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 5.481403582301, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 1.0811, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 5.9629615952333, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 1.0643, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 5.713885290476198, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0317, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 4.219414425585801, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.9974, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 4.065797123571697, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.9639, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 3.000906189449766, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.9135, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 2.6736043197666106, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.9189, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 2.331259637754972, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.8609, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.8307413373112864, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.8479, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 1.3386248131579794, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.8237, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.405386058044596, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.822, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.4553217181049751, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.8114, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 1.206012580100271, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.7602, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 1.0541117179863777, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.7715, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 1.0702812229320293, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.7711, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 1.0749271181633795, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.7636, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 1.0319747012953704, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.7409, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 1.2080822930552886, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.6265, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.7685179440440311, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.5932, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 1.0898022299054904, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.5744, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 1.1217749562547903, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.5856, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.9558838369638398, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.5833, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.8724388392568518, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.5746, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.833633108789087, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.5844, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.8018989761879962, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.5433, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.7401611867763375, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.5484, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.8070669670057865, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.5634, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.9072962519661446, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.4645, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.8250329895417012, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.4295, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.9405356181416574, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.4111, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.9163162294803638, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.4115, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.7848743270338966, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.3931, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.7641602580043291, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.3882, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.7029975068978057, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.3946, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.7222373191824814, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.4204, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.6981626946870054, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.3822, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.7233807530749774, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.3747, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.729461004858162, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.3757, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.823838779504481, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.3066, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.6005802004068651, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.3311, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.6210915339269238, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.3051, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.8086840737390982, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.3002, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.6813884177160316, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.2833, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.5322718048791336, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.2563, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.5800393319276762, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.2832, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.5907246820242783, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.2656, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.5989933418338111, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.2866, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.47406411778122354, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.2686, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.55294515604404, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.2992, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.5188492667294571, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.2236, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.4644799184184741, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.2247, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.531111726649738, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.252, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.4713995235775459, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.19, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.47715800330339564, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.2247, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.4615235804461352, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.2364, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.4704680713946454, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.2307, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.4081381943806433, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.2486, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.4286928393953709, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.2418, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.47139304404331345, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.1956, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.4216636227908017, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.221, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.42291930842614156, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.189, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.4210487751358509, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.2004, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.4271241205418503, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.2495, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.4207362803269532, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.2204, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.43160029278254547, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.2364, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.4100404936377231, | |
| "learning_rate": 0.0, | |
| "loss": 0.2309, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 8.131951929694618e+16, | |
| "train_loss": 0.5265720942190715, | |
| "train_runtime": 3256.0082, | |
| "train_samples_per_second": 2.15, | |
| "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": 8.131951929694618e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |