| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.666666666666667, | |
| "eval_steps": 500, | |
| "global_step": 70, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09523809523809523, | |
| "grad_norm": 6.345660437504658, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.8233, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": 6.283622045603061, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.8304, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 5.412658529172549, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.7657, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.38095238095238093, | |
| "grad_norm": 2.345915643260433, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.7137, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.47619047619047616, | |
| "grad_norm": 3.978245714271179, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.7344, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 4.109361336594632, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.7425, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 4.192469134634469, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6917, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.7619047619047619, | |
| "grad_norm": 2.5176185841550685, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.6805, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 2.328648019102948, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.6464, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 2.036921049465649, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.6255, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0476190476190477, | |
| "grad_norm": 2.069765632507996, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.9136, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 1.1896360139383995, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.5967, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.2380952380952381, | |
| "grad_norm": 1.0970057775113025, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.5809, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 0.945963241780214, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.5452, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 0.8647970492481037, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.5676, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.5238095238095237, | |
| "grad_norm": 0.8851693003981936, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.5428, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.619047619047619, | |
| "grad_norm": 0.7265497326291372, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.5498, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 0.737128766304549, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.5303, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8095238095238095, | |
| "grad_norm": 0.7826971377916968, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.5367, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 0.7209780278521523, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.5313, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.8982438092468968, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.7569, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.0952380952380953, | |
| "grad_norm": 0.6271449106805012, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.4821, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.1904761904761907, | |
| "grad_norm": 0.6572955894811892, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.4765, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 0.541780029737365, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.4771, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "grad_norm": 0.613586178276287, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.4765, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.4761904761904763, | |
| "grad_norm": 0.5948250757306962, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.491, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.571428571428571, | |
| "grad_norm": 0.4704705464037426, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.4672, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.5131546193421374, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.4672, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.761904761904762, | |
| "grad_norm": 0.4453558259513002, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.4643, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.5833856474638323, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.4584, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.9523809523809526, | |
| "grad_norm": 0.42491406444543783, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.4644, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.0476190476190474, | |
| "grad_norm": 0.6591926037742154, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.684, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.142857142857143, | |
| "grad_norm": 0.49442279927122396, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.4266, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.238095238095238, | |
| "grad_norm": 0.5157079270660057, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.431, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 0.37104599007265743, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.4191, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.4285714285714284, | |
| "grad_norm": 0.531573112084802, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.4497, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.5238095238095237, | |
| "grad_norm": 0.42068506755830964, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.3956, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.619047619047619, | |
| "grad_norm": 0.363336372372114, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.4056, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7142857142857144, | |
| "grad_norm": 0.4391710397027919, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.4245, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "grad_norm": 0.412424397736675, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.4401, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.9047619047619047, | |
| "grad_norm": 0.33643393975164165, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.4108, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.563992526685577, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.595, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.095238095238095, | |
| "grad_norm": 0.34690391329064163, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.4106, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.190476190476191, | |
| "grad_norm": 0.33682293614419584, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.3898, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 0.3390829928134016, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.3785, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.380952380952381, | |
| "grad_norm": 0.3840667477982577, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.3839, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.476190476190476, | |
| "grad_norm": 0.2987597342275811, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.3769, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "grad_norm": 0.29899372752471826, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.4029, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.666666666666667, | |
| "grad_norm": 0.3242145434252832, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.3745, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.761904761904762, | |
| "grad_norm": 0.2923410124766503, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.3697, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "grad_norm": 0.28177736516289853, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.3759, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.9523809523809526, | |
| "grad_norm": 0.287333903234355, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.3842, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.0476190476190474, | |
| "grad_norm": 0.4098097289280779, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.56, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.142857142857143, | |
| "grad_norm": 0.2394917075988088, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.3562, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.238095238095238, | |
| "grad_norm": 0.24945374095053138, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.3751, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.333333333333333, | |
| "grad_norm": 0.25561089906591966, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.3732, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.428571428571429, | |
| "grad_norm": 0.24030813696816458, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.3636, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.523809523809524, | |
| "grad_norm": 0.23622710776255834, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.3618, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.619047619047619, | |
| "grad_norm": 0.23647302619833374, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.3584, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 0.23867094974007025, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.3682, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.809523809523809, | |
| "grad_norm": 0.23238462116075423, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.3592, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.904761904761905, | |
| "grad_norm": 0.2337666373796116, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.3667, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.3592988182453475, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.5279, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.095238095238095, | |
| "grad_norm": 0.2174157808820532, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.3703, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.190476190476191, | |
| "grad_norm": 0.21854013953867216, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.3555, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.285714285714286, | |
| "grad_norm": 0.2248372779175572, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.3583, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.380952380952381, | |
| "grad_norm": 0.20416944156773195, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.3582, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.476190476190476, | |
| "grad_norm": 0.20203630226926828, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.353, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.571428571428571, | |
| "grad_norm": 0.20385260414575038, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.3689, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 0.20064643368826138, | |
| "learning_rate": 0.0, | |
| "loss": 0.3441, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "step": 70, | |
| "total_flos": 2.3541271679205376e+17, | |
| "train_loss": 0.4919714012316295, | |
| "train_runtime": 2750.5888, | |
| "train_samples_per_second": 2.545, | |
| "train_steps_per_second": 0.025 | |
| } | |
| ], | |
| "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": 2.3541271679205376e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |