| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.03474222015791918, | |
| "eval_steps": 17, | |
| "global_step": 187, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00018578727357176033, | |
| "eval_loss": 1.5152398347854614, | |
| "eval_runtime": 275.8856, | |
| "eval_samples_per_second": 32.861, | |
| "eval_steps_per_second": 4.11, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.000557361820715281, | |
| "grad_norm": 4.119506359100342, | |
| "learning_rate": 3e-05, | |
| "loss": 1.6013, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.001114723641430562, | |
| "grad_norm": 11.024518966674805, | |
| "learning_rate": 6e-05, | |
| "loss": 1.3784, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.001672085462145843, | |
| "grad_norm": 3.6604344844818115, | |
| "learning_rate": 9e-05, | |
| "loss": 1.2692, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.002229447282861124, | |
| "grad_norm": 12.112958908081055, | |
| "learning_rate": 9.997266286704631e-05, | |
| "loss": 1.1006, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.002786809103576405, | |
| "grad_norm": 2.718080997467041, | |
| "learning_rate": 9.98292246503335e-05, | |
| "loss": 1.1317, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0031583836507199257, | |
| "eval_loss": 0.9914608597755432, | |
| "eval_runtime": 278.075, | |
| "eval_samples_per_second": 32.603, | |
| "eval_steps_per_second": 4.078, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.003344170924291686, | |
| "grad_norm": 1.8613860607147217, | |
| "learning_rate": 9.956320346634876e-05, | |
| "loss": 0.994, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0039015327450069672, | |
| "grad_norm": 1.679516315460205, | |
| "learning_rate": 9.917525374361912e-05, | |
| "loss": 0.8489, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.004458894565722248, | |
| "grad_norm": 1.1957541704177856, | |
| "learning_rate": 9.86663298624003e-05, | |
| "loss": 0.951, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.005016256386437529, | |
| "grad_norm": 0.9221785068511963, | |
| "learning_rate": 9.803768380684242e-05, | |
| "loss": 0.7897, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.00557361820715281, | |
| "grad_norm": 1.2563328742980957, | |
| "learning_rate": 9.729086208503174e-05, | |
| "loss": 0.8182, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.006130980027868091, | |
| "grad_norm": 1.1113193035125732, | |
| "learning_rate": 9.642770192448536e-05, | |
| "loss": 0.8223, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.006316767301439851, | |
| "eval_loss": 0.8207408785820007, | |
| "eval_runtime": 278.0489, | |
| "eval_samples_per_second": 32.606, | |
| "eval_steps_per_second": 4.078, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.006688341848583372, | |
| "grad_norm": 1.2708672285079956, | |
| "learning_rate": 9.545032675245813e-05, | |
| "loss": 0.8516, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.007245703669298653, | |
| "grad_norm": 1.0124180316925049, | |
| "learning_rate": 9.43611409721806e-05, | |
| "loss": 0.8231, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0078030654900139345, | |
| "grad_norm": 0.8520745038986206, | |
| "learning_rate": 9.316282404787871e-05, | |
| "loss": 0.7552, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.008360427310729215, | |
| "grad_norm": 1.1972272396087646, | |
| "learning_rate": 9.185832391312644e-05, | |
| "loss": 0.7681, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.008917789131444497, | |
| "grad_norm": 0.9560966491699219, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.7997, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.009475150952159776, | |
| "grad_norm": 0.8837025165557861, | |
| "learning_rate": 8.894386393810563e-05, | |
| "loss": 0.7901, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.009475150952159776, | |
| "eval_loss": 0.7770994305610657, | |
| "eval_runtime": 278.1693, | |
| "eval_samples_per_second": 32.592, | |
| "eval_steps_per_second": 4.077, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.010032512772875058, | |
| "grad_norm": 1.023627519607544, | |
| "learning_rate": 8.73410738492077e-05, | |
| "loss": 0.7579, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.01058987459359034, | |
| "grad_norm": 0.8764439225196838, | |
| "learning_rate": 8.564642241456986e-05, | |
| "loss": 0.7208, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.01114723641430562, | |
| "grad_norm": 0.7949612140655518, | |
| "learning_rate": 8.386407858128706e-05, | |
| "loss": 0.7907, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0117045982350209, | |
| "grad_norm": 1.581484079360962, | |
| "learning_rate": 8.199842702516583e-05, | |
| "loss": 0.7894, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.012261960055736182, | |
| "grad_norm": 1.0892794132232666, | |
| "learning_rate": 8.005405736415126e-05, | |
| "loss": 0.8245, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.012633534602879703, | |
| "eval_loss": 0.7482102513313293, | |
| "eval_runtime": 278.4703, | |
| "eval_samples_per_second": 32.556, | |
| "eval_steps_per_second": 4.072, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.012819321876451463, | |
| "grad_norm": 0.9416138529777527, | |
| "learning_rate": 7.803575286758364e-05, | |
| "loss": 0.7452, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.013376683697166745, | |
| "grad_norm": 0.9427577257156372, | |
| "learning_rate": 7.594847868906076e-05, | |
| "loss": 0.7362, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.013934045517882025, | |
| "grad_norm": 0.8998479247093201, | |
| "learning_rate": 7.379736965185368e-05, | |
| "loss": 0.6653, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.014491407338597306, | |
| "grad_norm": 0.9068602919578552, | |
| "learning_rate": 7.158771761692464e-05, | |
| "loss": 0.7179, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.015048769159312587, | |
| "grad_norm": 0.8577080368995667, | |
| "learning_rate": 6.932495846462261e-05, | |
| "loss": 0.7525, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.015606130980027869, | |
| "grad_norm": 1.0340560674667358, | |
| "learning_rate": 6.701465872208216e-05, | |
| "loss": 0.7724, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.01579191825359963, | |
| "eval_loss": 0.7278433442115784, | |
| "eval_runtime": 278.2156, | |
| "eval_samples_per_second": 32.586, | |
| "eval_steps_per_second": 4.076, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.01616349280074315, | |
| "grad_norm": 0.8426377773284912, | |
| "learning_rate": 6.466250186922325e-05, | |
| "loss": 0.751, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.01672085462145843, | |
| "grad_norm": 0.6842407584190369, | |
| "learning_rate": 6.227427435703997e-05, | |
| "loss": 0.6239, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.01727821644217371, | |
| "grad_norm": 0.943516731262207, | |
| "learning_rate": 5.985585137257401e-05, | |
| "loss": 0.6827, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.017835578262888993, | |
| "grad_norm": 0.9946945905685425, | |
| "learning_rate": 5.74131823855921e-05, | |
| "loss": 0.6946, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.018392940083604273, | |
| "grad_norm": 0.9049971699714661, | |
| "learning_rate": 5.495227651252315e-05, | |
| "loss": 0.7255, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.018950301904319553, | |
| "grad_norm": 0.8010388016700745, | |
| "learning_rate": 5.247918773366112e-05, | |
| "loss": 0.7617, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.018950301904319553, | |
| "eval_loss": 0.7182794809341431, | |
| "eval_runtime": 278.3252, | |
| "eval_samples_per_second": 32.573, | |
| "eval_steps_per_second": 4.074, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.019507663725034836, | |
| "grad_norm": 0.8912818431854248, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6662, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.020065025545750115, | |
| "grad_norm": 0.8037520051002502, | |
| "learning_rate": 4.7520812266338885e-05, | |
| "loss": 0.655, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.0206223873664654, | |
| "grad_norm": 0.9647955894470215, | |
| "learning_rate": 4.504772348747687e-05, | |
| "loss": 0.7077, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 0.02117974918718068, | |
| "grad_norm": 0.7562959790229797, | |
| "learning_rate": 4.2586817614407895e-05, | |
| "loss": 0.6977, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 0.021737111007895958, | |
| "grad_norm": 0.7197637557983398, | |
| "learning_rate": 4.0144148627425993e-05, | |
| "loss": 0.7263, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.02210868555503948, | |
| "eval_loss": 0.7108221054077148, | |
| "eval_runtime": 278.2385, | |
| "eval_samples_per_second": 32.584, | |
| "eval_steps_per_second": 4.076, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 0.02229447282861124, | |
| "grad_norm": 0.7262760400772095, | |
| "learning_rate": 3.772572564296005e-05, | |
| "loss": 0.6754, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.02285183464932652, | |
| "grad_norm": 0.6029306650161743, | |
| "learning_rate": 3.533749813077677e-05, | |
| "loss": 0.7083, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 0.0234091964700418, | |
| "grad_norm": 0.7203994393348694, | |
| "learning_rate": 3.298534127791785e-05, | |
| "loss": 0.6682, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.023966558290757084, | |
| "grad_norm": 0.663855254650116, | |
| "learning_rate": 3.0675041535377405e-05, | |
| "loss": 0.6607, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 0.024523920111472364, | |
| "grad_norm": 1.1171367168426514, | |
| "learning_rate": 2.8412282383075363e-05, | |
| "loss": 0.7137, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.025081281932187643, | |
| "grad_norm": 0.8257487416267395, | |
| "learning_rate": 2.6202630348146324e-05, | |
| "loss": 0.6731, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.025267069205759406, | |
| "eval_loss": 0.7051795125007629, | |
| "eval_runtime": 278.2312, | |
| "eval_samples_per_second": 32.584, | |
| "eval_steps_per_second": 4.076, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.025638643752902927, | |
| "grad_norm": 0.7469737529754639, | |
| "learning_rate": 2.405152131093926e-05, | |
| "loss": 0.66, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 0.026196005573618206, | |
| "grad_norm": 0.7631163001060486, | |
| "learning_rate": 2.196424713241637e-05, | |
| "loss": 0.7156, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 0.02675336739433349, | |
| "grad_norm": 0.8524065017700195, | |
| "learning_rate": 1.9945942635848748e-05, | |
| "loss": 0.6498, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.02731072921504877, | |
| "grad_norm": 0.7686144113540649, | |
| "learning_rate": 1.800157297483417e-05, | |
| "loss": 0.6891, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 0.02786809103576405, | |
| "grad_norm": 0.8990215063095093, | |
| "learning_rate": 1.6135921418712956e-05, | |
| "loss": 0.583, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.028425452856479332, | |
| "grad_norm": 0.7455560564994812, | |
| "learning_rate": 1.435357758543015e-05, | |
| "loss": 0.6584, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 0.028425452856479332, | |
| "eval_loss": 0.7012729048728943, | |
| "eval_runtime": 278.4629, | |
| "eval_samples_per_second": 32.557, | |
| "eval_steps_per_second": 4.072, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 0.028982814677194612, | |
| "grad_norm": 0.7660591006278992, | |
| "learning_rate": 1.2658926150792322e-05, | |
| "loss": 0.6667, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.02954017649790989, | |
| "grad_norm": 0.6604006290435791, | |
| "learning_rate": 1.1056136061894384e-05, | |
| "loss": 0.6237, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 0.030097538318625175, | |
| "grad_norm": 0.8076897263526917, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.6899, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 0.030654900139340455, | |
| "grad_norm": 0.8138939738273621, | |
| "learning_rate": 8.141676086873572e-06, | |
| "loss": 0.606, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.031212261960055738, | |
| "grad_norm": 0.7922996878623962, | |
| "learning_rate": 6.837175952121306e-06, | |
| "loss": 0.7121, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.03158383650719926, | |
| "eval_loss": 0.6988500356674194, | |
| "eval_runtime": 278.3236, | |
| "eval_samples_per_second": 32.574, | |
| "eval_steps_per_second": 4.074, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.031769623780771014, | |
| "grad_norm": 0.8445965051651001, | |
| "learning_rate": 5.6388590278194096e-06, | |
| "loss": 0.6391, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 0.0323269856014863, | |
| "grad_norm": 0.7874613404273987, | |
| "learning_rate": 4.549673247541875e-06, | |
| "loss": 0.7674, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 0.03288434742220158, | |
| "grad_norm": 0.8437705636024475, | |
| "learning_rate": 3.5722980755146517e-06, | |
| "loss": 0.6699, | |
| "step": 177 | |
| }, | |
| { | |
| "epoch": 0.03344170924291686, | |
| "grad_norm": 0.7565687298774719, | |
| "learning_rate": 2.7091379149682685e-06, | |
| "loss": 0.7346, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.03399907106363214, | |
| "grad_norm": 0.7713425159454346, | |
| "learning_rate": 1.962316193157593e-06, | |
| "loss": 0.6421, | |
| "step": 183 | |
| }, | |
| { | |
| "epoch": 0.03455643288434742, | |
| "grad_norm": 0.755294144153595, | |
| "learning_rate": 1.333670137599713e-06, | |
| "loss": 0.6981, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 0.03474222015791918, | |
| "eval_loss": 0.6977519392967224, | |
| "eval_runtime": 278.5744, | |
| "eval_samples_per_second": 32.544, | |
| "eval_steps_per_second": 4.071, | |
| "step": 187 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 17, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.567372765580493e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |