| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.3362757461118117, | |
| "eval_steps": 50, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0016813787305590584, | |
| "eval_loss": 2.9795913696289062, | |
| "eval_runtime": 18.5959, | |
| "eval_samples_per_second": 53.883, | |
| "eval_steps_per_second": 6.776, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.008406893652795292, | |
| "grad_norm": 4.507114887237549, | |
| "learning_rate": 5e-05, | |
| "loss": 2.9614, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.016813787305590584, | |
| "grad_norm": 4.8396687507629395, | |
| "learning_rate": 0.0001, | |
| "loss": 2.557, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.025220680958385876, | |
| "grad_norm": 3.7924869060516357, | |
| "learning_rate": 9.98292246503335e-05, | |
| "loss": 1.8195, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.03362757461118117, | |
| "grad_norm": 3.164088487625122, | |
| "learning_rate": 9.931806517013612e-05, | |
| "loss": 1.6976, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.042034468263976464, | |
| "grad_norm": 2.6674678325653076, | |
| "learning_rate": 9.847001329696653e-05, | |
| "loss": 1.6413, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.05044136191677175, | |
| "grad_norm": 3.1234614849090576, | |
| "learning_rate": 9.729086208503174e-05, | |
| "loss": 1.659, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.05884825556956705, | |
| "grad_norm": 2.399535655975342, | |
| "learning_rate": 9.578866633275288e-05, | |
| "loss": 1.6538, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.06725514922236234, | |
| "grad_norm": 2.7914302349090576, | |
| "learning_rate": 9.397368756032445e-05, | |
| "loss": 1.6292, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.07566204287515763, | |
| "grad_norm": 2.3111212253570557, | |
| "learning_rate": 9.185832391312644e-05, | |
| "loss": 1.5684, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.08406893652795293, | |
| "grad_norm": 2.425607919692993, | |
| "learning_rate": 8.945702546981969e-05, | |
| "loss": 1.5452, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08406893652795293, | |
| "eval_loss": 1.5762242078781128, | |
| "eval_runtime": 18.8933, | |
| "eval_samples_per_second": 53.035, | |
| "eval_steps_per_second": 6.669, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.09247583018074822, | |
| "grad_norm": 2.81095552444458, | |
| "learning_rate": 8.678619553365659e-05, | |
| "loss": 1.6016, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.1008827238335435, | |
| "grad_norm": 2.1904377937316895, | |
| "learning_rate": 8.386407858128706e-05, | |
| "loss": 1.588, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.1092896174863388, | |
| "grad_norm": 2.673210859298706, | |
| "learning_rate": 8.07106356344834e-05, | |
| "loss": 1.4122, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.1176965111391341, | |
| "grad_norm": 2.258984327316284, | |
| "learning_rate": 7.734740790612136e-05, | |
| "loss": 1.5837, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.12610340479192939, | |
| "grad_norm": 1.992119312286377, | |
| "learning_rate": 7.379736965185368e-05, | |
| "loss": 1.489, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.13451029844472467, | |
| "grad_norm": 2.4318251609802246, | |
| "learning_rate": 7.008477123264848e-05, | |
| "loss": 1.6233, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.14291719209751996, | |
| "grad_norm": 2.4074137210845947, | |
| "learning_rate": 6.623497346023418e-05, | |
| "loss": 1.7463, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.15132408575031525, | |
| "grad_norm": 2.5759310722351074, | |
| "learning_rate": 6.227427435703997e-05, | |
| "loss": 1.4016, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.15973097940311054, | |
| "grad_norm": 2.0849971771240234, | |
| "learning_rate": 5.8229729514036705e-05, | |
| "loss": 1.4373, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.16813787305590586, | |
| "grad_norm": 2.3048596382141113, | |
| "learning_rate": 5.4128967273616625e-05, | |
| "loss": 1.567, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.16813787305590586, | |
| "eval_loss": 1.49704110622406, | |
| "eval_runtime": 18.708, | |
| "eval_samples_per_second": 53.56, | |
| "eval_steps_per_second": 6.735, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.17654476670870115, | |
| "grad_norm": 2.6232118606567383, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4021, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.18495166036149643, | |
| "grad_norm": 2.3169777393341064, | |
| "learning_rate": 4.5871032726383386e-05, | |
| "loss": 1.6127, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.19335855401429172, | |
| "grad_norm": 2.8697993755340576, | |
| "learning_rate": 4.17702704859633e-05, | |
| "loss": 1.3656, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.201765447667087, | |
| "grad_norm": 2.10636305809021, | |
| "learning_rate": 3.772572564296005e-05, | |
| "loss": 1.4455, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.2101723413198823, | |
| "grad_norm": 2.08447265625, | |
| "learning_rate": 3.3765026539765834e-05, | |
| "loss": 1.4558, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.2185792349726776, | |
| "grad_norm": 2.365922451019287, | |
| "learning_rate": 2.991522876735154e-05, | |
| "loss": 1.3913, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.22698612862547288, | |
| "grad_norm": 2.088407278060913, | |
| "learning_rate": 2.6202630348146324e-05, | |
| "loss": 1.617, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.2353930222782682, | |
| "grad_norm": 2.644561767578125, | |
| "learning_rate": 2.2652592093878666e-05, | |
| "loss": 1.3617, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.24379991593106348, | |
| "grad_norm": 2.082289457321167, | |
| "learning_rate": 1.928936436551661e-05, | |
| "loss": 1.3557, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.25220680958385877, | |
| "grad_norm": 2.023592710494995, | |
| "learning_rate": 1.6135921418712956e-05, | |
| "loss": 1.4909, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.25220680958385877, | |
| "eval_loss": 1.4728471040725708, | |
| "eval_runtime": 18.7448, | |
| "eval_samples_per_second": 53.455, | |
| "eval_steps_per_second": 6.722, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.26061370323665406, | |
| "grad_norm": 2.3685669898986816, | |
| "learning_rate": 1.3213804466343421e-05, | |
| "loss": 1.447, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.26902059688944935, | |
| "grad_norm": 2.5568618774414062, | |
| "learning_rate": 1.0542974530180327e-05, | |
| "loss": 1.5709, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.27742749054224464, | |
| "grad_norm": 2.6576414108276367, | |
| "learning_rate": 8.141676086873572e-06, | |
| "loss": 1.3218, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.2858343841950399, | |
| "grad_norm": 2.336989402770996, | |
| "learning_rate": 6.026312439675552e-06, | |
| "loss": 1.5184, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.2942412778478352, | |
| "grad_norm": 2.0917623043060303, | |
| "learning_rate": 4.2113336672471245e-06, | |
| "loss": 1.4752, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.3026481715006305, | |
| "grad_norm": 2.3651535511016846, | |
| "learning_rate": 2.7091379149682685e-06, | |
| "loss": 1.432, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.3110550651534258, | |
| "grad_norm": 2.6364834308624268, | |
| "learning_rate": 1.5299867030334814e-06, | |
| "loss": 1.516, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.3194619588062211, | |
| "grad_norm": 2.9864797592163086, | |
| "learning_rate": 6.819348298638839e-07, | |
| "loss": 1.3804, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.32786885245901637, | |
| "grad_norm": 2.326068878173828, | |
| "learning_rate": 1.7077534966650766e-07, | |
| "loss": 1.4848, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.3362757461118117, | |
| "grad_norm": 1.931203842163086, | |
| "learning_rate": 0.0, | |
| "loss": 1.4475, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3362757461118117, | |
| "eval_loss": 1.4650206565856934, | |
| "eval_runtime": 18.603, | |
| "eval_samples_per_second": 53.862, | |
| "eval_steps_per_second": 6.773, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 3.08713927737344e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |