| { | |
| "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": 6.7642642858010165, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.8748, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.258488353933081, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.8273, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 5.966080371671221, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.817, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.511604110884161, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.714, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 3.8366596467566487, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.7057, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.119577182176399, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.6928, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.783993055662112, | |
| "learning_rate": 2e-05, | |
| "loss": 0.678, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 3.6555781174397732, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.6895, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.938009730635268, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.6304, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 2.584718761745634, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.603, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 2.1879389957939823, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.944, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.389238091609866, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.5206, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.2195247721010125, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.5668, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 0.8007943168746146, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.525, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.165697470336033, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.5242, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 0.9196782417091595, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.5314, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.7462282494462885, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.4687, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.9320308329370559, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.6051, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.8611808583747746, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.5136, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.6647786884261824, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.4262, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 1.2496038331614818, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.8104, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.7387225372157543, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.4313, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.6491650785958907, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.4101, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.6220832082422753, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.4483, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.5474081025849494, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.4023, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.5324766780950909, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.419, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.616728214185472, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.4062, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.6619344277911274, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.4024, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.5505171779715399, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.4307, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.6088837691451525, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.3997, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.6930789695071727, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.4838, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.6934089393171236, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.4752, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.6118504930477825, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.3816, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.49659928371438694, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.3212, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.5997576579639787, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.3465, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.5577265847548161, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.3415, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.8241971193283669, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.3218, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.5251305814688372, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.339, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.5118900091927868, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.3344, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.49652352780926323, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.3109, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.5235550334137987, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.3157, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.8342202935711379, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.5153, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.41066936291648076, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.2417, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.5401757874993868, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.2689, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.42885925317032675, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.2403, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.5408626883358346, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.2863, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.5499459858203606, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.2979, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.48022074078812677, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.2829, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.4117973364831111, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.2276, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.4292045816006096, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.262, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.588945418300669, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.2595, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.611220303562635, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.3518, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.5930285967785092, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.2962, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.4855994803557773, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.2362, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.37295323220407345, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.2374, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.3988179533157459, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.2252, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.4376525527881068, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.2079, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.4418347597335303, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.223, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.42051287139894966, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.2256, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.3911655343812678, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.2153, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.3847056509072192, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.2422, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.3933846551975749, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.2212, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.6466746373552904, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.3673, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.3614376259184872, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.1977, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.35500312782473553, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.2356, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.3114926868537771, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.1885, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.36138535596832394, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.2322, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.2991637388947163, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.1829, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.35659792052061073, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.2222, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.3162119652698281, | |
| "learning_rate": 0.0, | |
| "loss": 0.203, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 9.868859925541683e+16, | |
| "train_loss": 0.41120058127811976, | |
| "train_runtime": 4376.5107, | |
| "train_samples_per_second": 1.599, | |
| "train_steps_per_second": 0.016 | |
| } | |
| ], | |
| "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": 9.868859925541683e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |