| { | |
| "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": 6.235518126009489, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.8044, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 5.997604186299459, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.7896, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 5.4658902644725575, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.7735, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.405270889628245, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.7148, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 3.8598645575572053, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.727, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.295206340047826, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.7267, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.095985587164726, | |
| "learning_rate": 2e-05, | |
| "loss": 0.661, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 2.6440352108839664, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.6753, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.4788966786501936, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.6668, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.8604098983659612, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.607, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 2.1663174615907517, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.9048, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.3089630942788186, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.5848, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.1321390284136819, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.5605, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 1.0267979951254338, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.5746, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.794202591167066, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.5125, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 0.9609625840755611, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.5524, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.7325220384574475, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.5138, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.8784782144311151, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.5446, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.615028889338724, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.5276, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.7839024782302995, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.5362, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 0.8390031062861036, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.7954, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.6671011053295518, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.5041, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.5894577535717058, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.4608, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.5800359230744153, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.4854, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.7524617321611401, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.4927, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.5590698277854202, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.4708, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.6363164554995804, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.4455, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.45673420525433656, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.4469, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.4515535774011796, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.4404, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.5641989686122225, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.4697, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.5293791731221562, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.5905, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.6374639516958273, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.5881, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.4903202868289181, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.4346, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.417197517454091, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.4028, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.4170670874232351, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.4287, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.4003124779130017, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.4021, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.4173097193902673, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.4286, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.3794874441152122, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.4216, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.35371543632141844, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.4019, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.37257246789921683, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.4162, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.36210419142351435, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.4211, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.5733044150502297, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.6319, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.2946266107976297, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.3793, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.3216060940320708, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.3706, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.35930481047497265, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.4009, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.30628293833684955, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.3847, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.29897729546482466, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.3859, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.3041478046795149, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.3693, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.3023377524958991, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.3876, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.27410677796379523, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.3756, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.2725638864738123, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.3644, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.46722898733778845, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.5816, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.29850549166088397, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.4086, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.23631061301058992, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.3693, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.23535014226875234, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.3427, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.26025801124632386, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.3656, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.24138785506732532, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.3644, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.2310544787475722, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.3491, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.2500449951412691, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.3814, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.27047900029698335, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.3629, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.23585039263565757, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.3583, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.20542158264550567, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.3556, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.36116253084769556, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.5569, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.19848209805276312, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.3394, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.20943734949469373, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.3623, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.2041407439636269, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.3497, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.20169222023423852, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.3612, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.20844309109598738, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.3535, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.21548571137992367, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.3518, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.20565882692085535, | |
| "learning_rate": 0.0, | |
| "loss": 0.3611, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 2.7539459660998246e+17, | |
| "train_loss": 0.48901969535010203, | |
| "train_runtime": 8393.0583, | |
| "train_samples_per_second": 0.834, | |
| "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": 2.7539459660998246e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |