| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 35, | |
| "global_step": 350, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.014285714285714285, | |
| "grad_norm": 118.65539307904277, | |
| "learning_rate": 1.6e-06, | |
| "loss": 0.8722, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.02857142857142857, | |
| "grad_norm": 38.484858029114484, | |
| "learning_rate": 3.2e-06, | |
| "loss": 0.7825, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04285714285714286, | |
| "grad_norm": 88.4372189754248, | |
| "learning_rate": 4.8e-06, | |
| "loss": 0.7646, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.05714285714285714, | |
| "grad_norm": 41.455396892104744, | |
| "learning_rate": 6.4e-06, | |
| "loss": 0.7226, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.07142857142857142, | |
| "grad_norm": 4.738758549704674, | |
| "learning_rate": 8e-06, | |
| "loss": 0.6666, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.08571428571428572, | |
| "grad_norm": 10.5176134179179, | |
| "learning_rate": 7.771005917159763e-06, | |
| "loss": 0.7263, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 11.488275336430554, | |
| "learning_rate": 7.545562130177514e-06, | |
| "loss": 0.6211, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 0.9106850624084473, | |
| "eval_runtime": 8.373, | |
| "eval_samples_per_second": 26.275, | |
| "eval_steps_per_second": 4.419, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.11428571428571428, | |
| "grad_norm": 18.92714653332412, | |
| "learning_rate": 7.323668639053254e-06, | |
| "loss": 0.6989, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.12857142857142856, | |
| "grad_norm": 22.493765084792706, | |
| "learning_rate": 7.105325443786982e-06, | |
| "loss": 0.704, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.14285714285714285, | |
| "grad_norm": 211.59677798517507, | |
| "learning_rate": 6.890532544378699e-06, | |
| "loss": 0.6654, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.15714285714285714, | |
| "grad_norm": 160.57396424563427, | |
| "learning_rate": 6.679289940828402e-06, | |
| "loss": 0.7127, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.17142857142857143, | |
| "grad_norm": 8.199790596273626, | |
| "learning_rate": 6.471597633136094e-06, | |
| "loss": 0.6253, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.18571428571428572, | |
| "grad_norm": 50.8343087092876, | |
| "learning_rate": 6.2674556213017745e-06, | |
| "loss": 0.6973, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 99.89969831226269, | |
| "learning_rate": 6.066863905325444e-06, | |
| "loss": 0.6976, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.852588951587677, | |
| "eval_runtime": 8.1201, | |
| "eval_samples_per_second": 27.093, | |
| "eval_steps_per_second": 4.557, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.21428571428571427, | |
| "grad_norm": 68.18379393534644, | |
| "learning_rate": 5.869822485207101e-06, | |
| "loss": 0.6194, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.22857142857142856, | |
| "grad_norm": 10.735154673670019, | |
| "learning_rate": 5.676331360946745e-06, | |
| "loss": 0.6906, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.24285714285714285, | |
| "grad_norm": 9.131803561236161, | |
| "learning_rate": 5.486390532544378e-06, | |
| "loss": 0.6401, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.2571428571428571, | |
| "grad_norm": 3.7707612659287766, | |
| "learning_rate": 5.300000000000002e-06, | |
| "loss": 0.6326, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.2714285714285714, | |
| "grad_norm": 8.70600675802851, | |
| "learning_rate": 5.1171597633136094e-06, | |
| "loss": 0.6019, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 7.499128841456364, | |
| "learning_rate": 4.9378698224852065e-06, | |
| "loss": 0.6404, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 5.203399746039487, | |
| "learning_rate": 4.762130177514793e-06, | |
| "loss": 0.5762, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.7999083399772644, | |
| "eval_runtime": 8.0634, | |
| "eval_samples_per_second": 27.284, | |
| "eval_steps_per_second": 4.589, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.3142857142857143, | |
| "grad_norm": 4.900592118020277, | |
| "learning_rate": 4.589940828402367e-06, | |
| "loss": 0.7544, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.32857142857142857, | |
| "grad_norm": 6.583803116646508, | |
| "learning_rate": 4.421301775147928e-06, | |
| "loss": 0.6494, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.34285714285714286, | |
| "grad_norm": 2.8434894878977968, | |
| "learning_rate": 4.2562130177514784e-06, | |
| "loss": 0.6542, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.35714285714285715, | |
| "grad_norm": 3.3996307039750926, | |
| "learning_rate": 4.094674556213017e-06, | |
| "loss": 0.6631, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.37142857142857144, | |
| "grad_norm": 5.094034483082871, | |
| "learning_rate": 3.936686390532545e-06, | |
| "loss": 0.6086, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.38571428571428573, | |
| "grad_norm": 3.8103990086460757, | |
| "learning_rate": 3.7822485207100586e-06, | |
| "loss": 0.5935, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 4.098923421731261, | |
| "learning_rate": 3.631360946745561e-06, | |
| "loss": 0.5226, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.7717307209968567, | |
| "eval_runtime": 8.04, | |
| "eval_samples_per_second": 27.363, | |
| "eval_steps_per_second": 4.602, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.4142857142857143, | |
| "grad_norm": 3.3521626999923875, | |
| "learning_rate": 3.484023668639053e-06, | |
| "loss": 0.5967, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.42857142857142855, | |
| "grad_norm": 4.761409736989179, | |
| "learning_rate": 3.3402366863905327e-06, | |
| "loss": 0.6379, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.44285714285714284, | |
| "grad_norm": 5.189823443764368, | |
| "learning_rate": 3.1999999999999994e-06, | |
| "loss": 0.69, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.45714285714285713, | |
| "grad_norm": 3.4259629095344963, | |
| "learning_rate": 3.0633136094674547e-06, | |
| "loss": 0.6177, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.4714285714285714, | |
| "grad_norm": 6.564952365844063, | |
| "learning_rate": 2.930177514792899e-06, | |
| "loss": 0.6451, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.4857142857142857, | |
| "grad_norm": 3.8605502771298346, | |
| "learning_rate": 2.8005917159763313e-06, | |
| "loss": 0.5706, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 4.052127550985865, | |
| "learning_rate": 2.674556213017751e-06, | |
| "loss": 0.5866, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.7432886958122253, | |
| "eval_runtime": 8.0548, | |
| "eval_samples_per_second": 27.313, | |
| "eval_steps_per_second": 4.594, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.5142857142857142, | |
| "grad_norm": 3.3881437035921675, | |
| "learning_rate": 2.5520710059171586e-06, | |
| "loss": 0.6236, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.5285714285714286, | |
| "grad_norm": 3.6324500058620783, | |
| "learning_rate": 2.4331360946745558e-06, | |
| "loss": 0.614, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.5428571428571428, | |
| "grad_norm": 3.6721733069689377, | |
| "learning_rate": 2.317751479289941e-06, | |
| "loss": 0.5696, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.5571428571428572, | |
| "grad_norm": 3.2065483170410944, | |
| "learning_rate": 2.2059171597633135e-06, | |
| "loss": 0.6017, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 3.694107283298511, | |
| "learning_rate": 2.0976331360946745e-06, | |
| "loss": 0.5215, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5857142857142857, | |
| "grad_norm": 3.3237350491778606, | |
| "learning_rate": 1.9928994082840233e-06, | |
| "loss": 0.6335, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 5.7877669531548595, | |
| "learning_rate": 1.891715976331361e-06, | |
| "loss": 0.5999, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 0.7398412227630615, | |
| "eval_runtime": 8.0412, | |
| "eval_samples_per_second": 27.359, | |
| "eval_steps_per_second": 4.601, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6142857142857143, | |
| "grad_norm": 4.758098459635445, | |
| "learning_rate": 1.7940828402366863e-06, | |
| "loss": 0.6837, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.6285714285714286, | |
| "grad_norm": 2.672565209825915, | |
| "learning_rate": 1.7000000000000002e-06, | |
| "loss": 0.6247, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.6428571428571429, | |
| "grad_norm": 3.9151505842097594, | |
| "learning_rate": 1.6094674556213014e-06, | |
| "loss": 0.5437, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.6571428571428571, | |
| "grad_norm": 4.93388663600263, | |
| "learning_rate": 1.5224852071005916e-06, | |
| "loss": 0.6034, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.6714285714285714, | |
| "grad_norm": 2.9614002133596316, | |
| "learning_rate": 1.4390532544378696e-06, | |
| "loss": 0.5687, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.6857142857142857, | |
| "grad_norm": 3.738349914157848, | |
| "learning_rate": 1.3591715976331362e-06, | |
| "loss": 0.6636, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 3.8772829249948852, | |
| "learning_rate": 1.2828402366863903e-06, | |
| "loss": 0.6464, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_loss": 0.7203609943389893, | |
| "eval_runtime": 8.0382, | |
| "eval_samples_per_second": 27.369, | |
| "eval_steps_per_second": 4.603, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.7142857142857143, | |
| "grad_norm": 2.972054074443536, | |
| "learning_rate": 1.2100591715976333e-06, | |
| "loss": 0.6081, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.7285714285714285, | |
| "grad_norm": 2.8758679138381202, | |
| "learning_rate": 1.1408284023668636e-06, | |
| "loss": 0.5808, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.7428571428571429, | |
| "grad_norm": 3.959990245370596, | |
| "learning_rate": 1.0751479289940828e-06, | |
| "loss": 0.6201, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.7571428571428571, | |
| "grad_norm": 3.5665367608826632, | |
| "learning_rate": 1.0130177514792898e-06, | |
| "loss": 0.5034, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.7714285714285715, | |
| "grad_norm": 4.028636856853669, | |
| "learning_rate": 9.544378698224853e-07, | |
| "loss": 0.5932, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.7857142857142857, | |
| "grad_norm": 4.258016842155138, | |
| "learning_rate": 8.994082840236684e-07, | |
| "loss": 0.5742, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 3.5334205284517024, | |
| "learning_rate": 8.479289940828401e-07, | |
| "loss": 0.552, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.7106707692146301, | |
| "eval_runtime": 8.068, | |
| "eval_samples_per_second": 27.268, | |
| "eval_steps_per_second": 4.586, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.8142857142857143, | |
| "grad_norm": 6.70792314615095, | |
| "learning_rate": 7.999999999999998e-07, | |
| "loss": 0.6199, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.8285714285714286, | |
| "grad_norm": 3.3423971423196983, | |
| "learning_rate": 7.556213017751479e-07, | |
| "loss": 0.5638, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.8428571428571429, | |
| "grad_norm": 3.0750783422793497, | |
| "learning_rate": 7.147928994082838e-07, | |
| "loss": 0.5326, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 4.7296945533927355, | |
| "learning_rate": 6.775147928994083e-07, | |
| "loss": 0.6127, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.8714285714285714, | |
| "grad_norm": 3.833725884188028, | |
| "learning_rate": 6.437869822485206e-07, | |
| "loss": 0.5634, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.8857142857142857, | |
| "grad_norm": 3.359034703004336, | |
| "learning_rate": 6.136094674556213e-07, | |
| "loss": 0.5592, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 3.1893425724586857, | |
| "learning_rate": 5.869822485207099e-07, | |
| "loss": 0.5657, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_loss": 0.7130246758460999, | |
| "eval_runtime": 8.0419, | |
| "eval_samples_per_second": 27.357, | |
| "eval_steps_per_second": 4.601, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.9142857142857143, | |
| "grad_norm": 3.3324510082581087, | |
| "learning_rate": 5.63905325443787e-07, | |
| "loss": 0.6271, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.9285714285714286, | |
| "grad_norm": 3.4362159022422736, | |
| "learning_rate": 5.443786982248519e-07, | |
| "loss": 0.5349, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.9428571428571428, | |
| "grad_norm": 4.804926939525648, | |
| "learning_rate": 5.284023668639053e-07, | |
| "loss": 0.7169, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.9571428571428572, | |
| "grad_norm": 4.408287391472689, | |
| "learning_rate": 5.159763313609466e-07, | |
| "loss": 0.606, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.9714285714285714, | |
| "grad_norm": 5.066493752587581, | |
| "learning_rate": 5.071005917159763e-07, | |
| "loss": 0.5669, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.9857142857142858, | |
| "grad_norm": 8.178486160006328, | |
| "learning_rate": 5.01775147928994e-07, | |
| "loss": 0.5541, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 3.7353925732066693, | |
| "learning_rate": 5e-07, | |
| "loss": 0.5687, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.7043666839599609, | |
| "eval_runtime": 8.0404, | |
| "eval_samples_per_second": 27.362, | |
| "eval_steps_per_second": 4.602, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 350, | |
| "total_flos": 2.093861581933773e+16, | |
| "train_loss": 0.6267224325452532, | |
| "train_runtime": 1749.357, | |
| "train_samples_per_second": 3.601, | |
| "train_steps_per_second": 0.2 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 350, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 350, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.093861581933773e+16, | |
| "train_batch_size": 3, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |