| { | |
| "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": 6.30567489447412, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.8043, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.532654339441336, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.8279, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 5.7685132527404175, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.8038, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.410942839638782, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.7329, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 3.8915558589017385, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.7287, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.409796167724789, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.7545, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.367857170838888, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7052, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 2.6433956652473625, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.6657, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.4090427025015084, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.6227, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 2.1048578712596644, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.627, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 2.1303389267192547, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.9266, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.2492175909037224, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.6115, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.2063084923544758, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.5877, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 0.9656693683340709, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.5778, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.7700476606403841, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.528, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 0.9216952071561254, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.5606, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.7653456264604785, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.5509, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.7230605843223131, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.491, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.7957579383960275, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.5116, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.7532702180276306, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.5288, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 1.1356841519104781, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.8504, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.5971153673396078, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.5018, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.6586354992361362, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.4784, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.5680472893432346, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.4745, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.7350970938514598, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.4606, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.5323413360855892, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.4753, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.7346047973622335, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.4715, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.44211175573359096, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.4689, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.5569389749544316, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.4682, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.4929033591528241, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.4771, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.5792336722403362, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.5823, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.6313106265649959, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.6008, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.40910876360263554, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.4189, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.449210633855839, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.4247, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.4167983174243858, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.4253, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.4380863731547915, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.4318, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.4366226925192318, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.4243, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.42654765739146805, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.4039, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.3411401386511929, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.4299, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.34311832436615763, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.4257, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.33984800348060135, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.4011, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.6110266518937262, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.6705, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.3439873480034602, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.3922, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.33825897057042076, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.378, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.4213524972644932, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.3978, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.3522510180453797, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.3776, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.3008693165484971, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.38, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.31345609912051453, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.3727, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.34920170602454637, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.3813, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.30992564996288374, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.3926, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.2708538219740301, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.3883, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.42017157056178744, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.5637, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.3144304067082361, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.4025, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.2677378343577208, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.3831, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.24673532889638494, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.362, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.26011859613172966, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.3703, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.2605335921844092, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.3742, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.23917369843881595, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.3527, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.24762126594962444, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.3561, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.2543314163925074, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.371, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.22296821542170087, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.346, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.22886255450990947, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.3692, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.3805254302512643, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.5867, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.2072208900448251, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.3446, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.21205243251987602, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.3638, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.20734369132378783, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.3631, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.2101792350789993, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.3626, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.2172331308291079, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.3607, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.20426385395695218, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.3664, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.20809892448851816, | |
| "learning_rate": 0.0, | |
| "loss": 0.3444, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 2.7335868856192205e+17, | |
| "train_loss": 0.495952639409474, | |
| "train_runtime": 8363.9558, | |
| "train_samples_per_second": 0.837, | |
| "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": 2.7335868856192205e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |