| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.010204081632653, | |
| "eval_steps": 9, | |
| "global_step": 99, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01020408163265306, | |
| "eval_loss": 2.5547358989715576, | |
| "eval_runtime": 10.9643, | |
| "eval_samples_per_second": 15.049, | |
| "eval_steps_per_second": 1.915, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.030612244897959183, | |
| "grad_norm": 22.481250762939453, | |
| "learning_rate": 3e-05, | |
| "loss": 10.7284, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.061224489795918366, | |
| "grad_norm": 8.63905143737793, | |
| "learning_rate": 6e-05, | |
| "loss": 9.5021, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.09183673469387756, | |
| "grad_norm": 6.093153953552246, | |
| "learning_rate": 9e-05, | |
| "loss": 8.4335, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.09183673469387756, | |
| "eval_loss": 1.867263674736023, | |
| "eval_runtime": 11.1356, | |
| "eval_samples_per_second": 14.817, | |
| "eval_steps_per_second": 1.886, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.12244897959183673, | |
| "grad_norm": 5.729458332061768, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 7.2274, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.15306122448979592, | |
| "grad_norm": 5.534293174743652, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 7.2317, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1836734693877551, | |
| "grad_norm": 6.290515899658203, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 7.4009, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.1836734693877551, | |
| "eval_loss": 1.7399932146072388, | |
| "eval_runtime": 11.152, | |
| "eval_samples_per_second": 14.796, | |
| "eval_steps_per_second": 1.883, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.21428571428571427, | |
| "grad_norm": 5.591012477874756, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 7.2483, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.24489795918367346, | |
| "grad_norm": 4.81268310546875, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 7.0443, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.2755102040816326, | |
| "grad_norm": 4.390398025512695, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 6.7129, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.2755102040816326, | |
| "eval_loss": 1.676879644393921, | |
| "eval_runtime": 11.1798, | |
| "eval_samples_per_second": 14.759, | |
| "eval_steps_per_second": 1.878, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.30612244897959184, | |
| "grad_norm": 4.704708099365234, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 6.8546, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.336734693877551, | |
| "grad_norm": 4.343167304992676, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 6.9952, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.3673469387755102, | |
| "grad_norm": 3.5825247764587402, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 6.6506, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.3673469387755102, | |
| "eval_loss": 1.6432121992111206, | |
| "eval_runtime": 11.181, | |
| "eval_samples_per_second": 14.757, | |
| "eval_steps_per_second": 1.878, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.3979591836734694, | |
| "grad_norm": 4.151194095611572, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 6.538, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.42857142857142855, | |
| "grad_norm": 4.192036151885986, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 6.6947, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.45918367346938777, | |
| "grad_norm": 4.50932502746582, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 6.848, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.45918367346938777, | |
| "eval_loss": 1.622291922569275, | |
| "eval_runtime": 11.1576, | |
| "eval_samples_per_second": 14.788, | |
| "eval_steps_per_second": 1.882, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.4897959183673469, | |
| "grad_norm": 3.9591734409332275, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 6.6632, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.5204081632653061, | |
| "grad_norm": 3.3937931060791016, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 6.5888, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.5510204081632653, | |
| "grad_norm": 3.470456123352051, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 6.5541, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.5510204081632653, | |
| "eval_loss": 1.6126476526260376, | |
| "eval_runtime": 11.1832, | |
| "eval_samples_per_second": 14.754, | |
| "eval_steps_per_second": 1.878, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.5816326530612245, | |
| "grad_norm": 3.859346389770508, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 6.6553, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.6122448979591837, | |
| "grad_norm": 3.1429214477539062, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 6.4766, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.6428571428571429, | |
| "grad_norm": 4.059041500091553, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 6.416, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.6428571428571429, | |
| "eval_loss": 1.5922267436981201, | |
| "eval_runtime": 11.174, | |
| "eval_samples_per_second": 14.766, | |
| "eval_steps_per_second": 1.879, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.673469387755102, | |
| "grad_norm": 4.152980804443359, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 6.4115, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.7040816326530612, | |
| "grad_norm": 3.4916810989379883, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 6.4901, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.7346938775510204, | |
| "grad_norm": 3.7989633083343506, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 6.6644, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.7346938775510204, | |
| "eval_loss": 1.5862826108932495, | |
| "eval_runtime": 11.1677, | |
| "eval_samples_per_second": 14.775, | |
| "eval_steps_per_second": 1.88, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.7653061224489796, | |
| "grad_norm": 3.602689504623413, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 6.447, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.7959183673469388, | |
| "grad_norm": 3.743265390396118, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 6.4277, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.826530612244898, | |
| "grad_norm": 3.567052125930786, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 6.7132, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.826530612244898, | |
| "eval_loss": 1.5810420513153076, | |
| "eval_runtime": 11.1521, | |
| "eval_samples_per_second": 14.795, | |
| "eval_steps_per_second": 1.883, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 3.60418438911438, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 6.3228, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.8877551020408163, | |
| "grad_norm": 3.443751573562622, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 6.6778, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.9183673469387755, | |
| "grad_norm": 4.735818386077881, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 6.7016, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.9183673469387755, | |
| "eval_loss": 1.5796834230422974, | |
| "eval_runtime": 11.1726, | |
| "eval_samples_per_second": 14.768, | |
| "eval_steps_per_second": 1.88, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.9489795918367347, | |
| "grad_norm": 3.8432347774505615, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 6.5667, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.9795918367346939, | |
| "grad_norm": 3.743124008178711, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 6.4069, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.010204081632653, | |
| "grad_norm": 3.2074642181396484, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 6.4376, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.010204081632653, | |
| "eval_loss": 1.579128384590149, | |
| "eval_runtime": 11.1546, | |
| "eval_samples_per_second": 14.792, | |
| "eval_steps_per_second": 1.883, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.43458510209024e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |