| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 5, | |
| "global_step": 46, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.021739130434782608, | |
| "grad_norm": 2.5425055027008057, | |
| "learning_rate": 1e-05, | |
| "loss": 0.706, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.043478260869565216, | |
| "grad_norm": 2.3038032054901123, | |
| "learning_rate": 9.987820251299121e-06, | |
| "loss": 0.643, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.06521739130434782, | |
| "grad_norm": 2.222012996673584, | |
| "learning_rate": 9.951340343707852e-06, | |
| "loss": 0.722, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.08695652173913043, | |
| "grad_norm": 1.4986803531646729, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 0.6543, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.10869565217391304, | |
| "grad_norm": 1.5336999893188477, | |
| "learning_rate": 9.806308479691595e-06, | |
| "loss": 0.766, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.10869565217391304, | |
| "eval_loss": 0.5912319421768188, | |
| "eval_runtime": 0.8443, | |
| "eval_samples_per_second": 4.737, | |
| "eval_steps_per_second": 1.184, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.13043478260869565, | |
| "grad_norm": 1.4662328958511353, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 0.5495, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.15217391304347827, | |
| "grad_norm": 1.3746005296707153, | |
| "learning_rate": 9.567727288213005e-06, | |
| "loss": 0.5193, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.17391304347826086, | |
| "grad_norm": 1.4526853561401367, | |
| "learning_rate": 9.414737964294636e-06, | |
| "loss": 0.5578, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.1956521739130435, | |
| "grad_norm": 0.8964347243309021, | |
| "learning_rate": 9.24024048078213e-06, | |
| "loss": 0.3643, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.21739130434782608, | |
| "grad_norm": 1.2126158475875854, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 0.5873, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.21739130434782608, | |
| "eval_loss": 0.5282274484634399, | |
| "eval_runtime": 0.843, | |
| "eval_samples_per_second": 4.745, | |
| "eval_steps_per_second": 1.186, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2391304347826087, | |
| "grad_norm": 1.2183283567428589, | |
| "learning_rate": 8.83022221559489e-06, | |
| "loss": 0.6398, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.2608695652173913, | |
| "grad_norm": 0.9250560402870178, | |
| "learning_rate": 8.596699001693257e-06, | |
| "loss": 0.4296, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.2826086956521739, | |
| "grad_norm": 1.0050208568572998, | |
| "learning_rate": 8.345653031794292e-06, | |
| "loss": 0.5244, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.30434782608695654, | |
| "grad_norm": 1.0231624841690063, | |
| "learning_rate": 8.078307376628292e-06, | |
| "loss": 0.4739, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.32608695652173914, | |
| "grad_norm": 0.8328154683113098, | |
| "learning_rate": 7.795964517353734e-06, | |
| "loss": 0.3868, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.32608695652173914, | |
| "eval_loss": 0.49576932191848755, | |
| "eval_runtime": 0.8436, | |
| "eval_samples_per_second": 4.742, | |
| "eval_steps_per_second": 1.185, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 1.0239394903182983, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.5849, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.3695652173913043, | |
| "grad_norm": 0.9168555736541748, | |
| "learning_rate": 7.191855733945388e-06, | |
| "loss": 0.4854, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.391304347826087, | |
| "grad_norm": 0.9247157573699951, | |
| "learning_rate": 6.873032967079562e-06, | |
| "loss": 0.3887, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.41304347826086957, | |
| "grad_norm": 1.1805756092071533, | |
| "learning_rate": 6.545084971874738e-06, | |
| "loss": 0.6612, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.43478260869565216, | |
| "grad_norm": 0.9535987377166748, | |
| "learning_rate": 6.209609477998339e-06, | |
| "loss": 0.5101, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.43478260869565216, | |
| "eval_loss": 0.4761270582675934, | |
| "eval_runtime": 0.8437, | |
| "eval_samples_per_second": 4.741, | |
| "eval_steps_per_second": 1.185, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.45652173913043476, | |
| "grad_norm": 0.821264922618866, | |
| "learning_rate": 5.8682408883346535e-06, | |
| "loss": 0.4696, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.4782608695652174, | |
| "grad_norm": 0.9166697263717651, | |
| "learning_rate": 5.522642316338268e-06, | |
| "loss": 0.4555, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.7674450278282166, | |
| "learning_rate": 5.174497483512506e-06, | |
| "loss": 0.4064, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.5217391304347826, | |
| "grad_norm": 0.7370434999465942, | |
| "learning_rate": 4.825502516487497e-06, | |
| "loss": 0.3378, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5434782608695652, | |
| "grad_norm": 0.8126187324523926, | |
| "learning_rate": 4.477357683661734e-06, | |
| "loss": 0.4085, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.5434782608695652, | |
| "eval_loss": 0.46437764167785645, | |
| "eval_runtime": 0.8435, | |
| "eval_samples_per_second": 4.742, | |
| "eval_steps_per_second": 1.186, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.5652173913043478, | |
| "grad_norm": 0.8102895021438599, | |
| "learning_rate": 4.131759111665349e-06, | |
| "loss": 0.4565, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.5869565217391305, | |
| "grad_norm": 1.0004063844680786, | |
| "learning_rate": 3.790390522001662e-06, | |
| "loss": 0.613, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.6086956521739131, | |
| "grad_norm": 0.8794491291046143, | |
| "learning_rate": 3.4549150281252635e-06, | |
| "loss": 0.4919, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.6304347826086957, | |
| "grad_norm": 0.7904605865478516, | |
| "learning_rate": 3.12696703292044e-06, | |
| "loss": 0.4456, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.6521739130434783, | |
| "grad_norm": 0.8982527256011963, | |
| "learning_rate": 2.8081442660546126e-06, | |
| "loss": 0.5561, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.6521739130434783, | |
| "eval_loss": 0.45777273178100586, | |
| "eval_runtime": 0.8438, | |
| "eval_samples_per_second": 4.741, | |
| "eval_steps_per_second": 1.185, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.6739130434782609, | |
| "grad_norm": 0.7891985774040222, | |
| "learning_rate": 2.5000000000000015e-06, | |
| "loss": 0.4395, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 0.7818904519081116, | |
| "learning_rate": 2.204035482646267e-06, | |
| "loss": 0.4492, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.717391304347826, | |
| "grad_norm": 0.8261966705322266, | |
| "learning_rate": 1.9216926233717087e-06, | |
| "loss": 0.389, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.7391304347826086, | |
| "grad_norm": 0.8626196384429932, | |
| "learning_rate": 1.6543469682057105e-06, | |
| "loss": 0.4336, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.7608695652173914, | |
| "grad_norm": 0.8142871856689453, | |
| "learning_rate": 1.4033009983067454e-06, | |
| "loss": 0.4683, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.7608695652173914, | |
| "eval_loss": 0.45417019724845886, | |
| "eval_runtime": 0.8436, | |
| "eval_samples_per_second": 4.742, | |
| "eval_steps_per_second": 1.185, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.782608695652174, | |
| "grad_norm": 0.8224750757217407, | |
| "learning_rate": 1.1697777844051105e-06, | |
| "loss": 0.4277, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.8043478260869565, | |
| "grad_norm": 1.101127028465271, | |
| "learning_rate": 9.549150281252633e-07, | |
| "loss": 0.4057, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.8260869565217391, | |
| "grad_norm": 0.9483347535133362, | |
| "learning_rate": 7.597595192178702e-07, | |
| "loss": 0.5928, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.8478260869565217, | |
| "grad_norm": 0.9000388383865356, | |
| "learning_rate": 5.852620357053651e-07, | |
| "loss": 0.5955, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "grad_norm": 0.8246148228645325, | |
| "learning_rate": 4.322727117869951e-07, | |
| "loss": 0.5055, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "eval_loss": 0.452594518661499, | |
| "eval_runtime": 0.8425, | |
| "eval_samples_per_second": 4.748, | |
| "eval_steps_per_second": 1.187, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.8913043478260869, | |
| "grad_norm": 0.7705091834068298, | |
| "learning_rate": 3.015368960704584e-07, | |
| "loss": 0.4106, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.9130434782608695, | |
| "grad_norm": 0.7705535888671875, | |
| "learning_rate": 1.9369152030840553e-07, | |
| "loss": 0.4183, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.9347826086956522, | |
| "grad_norm": 0.78244948387146, | |
| "learning_rate": 1.0926199633097156e-07, | |
| "loss": 0.4416, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.9565217391304348, | |
| "grad_norm": 0.8442041277885437, | |
| "learning_rate": 4.865965629214819e-08, | |
| "loss": 0.4899, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.9782608695652174, | |
| "grad_norm": 0.8914652466773987, | |
| "learning_rate": 1.2179748700879013e-08, | |
| "loss": 0.5359, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.9782608695652174, | |
| "eval_loss": 0.45189881324768066, | |
| "eval_runtime": 0.843, | |
| "eval_samples_per_second": 4.745, | |
| "eval_steps_per_second": 1.186, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.797126829624176, | |
| "learning_rate": 0.0, | |
| "loss": 0.4124, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 46, | |
| "total_flos": 7.03161217014825e+16, | |
| "train_loss": 0.50457603516786, | |
| "train_runtime": 362.0803, | |
| "train_samples_per_second": 1.016, | |
| "train_steps_per_second": 0.127 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 46, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 46, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.03161217014825e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |