| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.7079303675048356, | |
| "eval_steps": 50, | |
| "global_step": 1400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09671179883945841, | |
| "grad_norm": 0.035103704780340195, | |
| "learning_rate": 0.00016025641025641028, | |
| "loss": 1.0483, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.09671179883945841, | |
| "eval_loss": 0.9260319471359253, | |
| "eval_runtime": 60.9012, | |
| "eval_samples_per_second": 1.642, | |
| "eval_steps_per_second": 0.213, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.19342359767891681, | |
| "grad_norm": 0.030684856697916985, | |
| "learning_rate": 0.00032051282051282057, | |
| "loss": 0.8577, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.19342359767891681, | |
| "eval_loss": 0.820220410823822, | |
| "eval_runtime": 56.5241, | |
| "eval_samples_per_second": 1.769, | |
| "eval_steps_per_second": 0.23, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2901353965183752, | |
| "grad_norm": 0.02802114002406597, | |
| "learning_rate": 0.0004807692307692308, | |
| "loss": 0.7996, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.2901353965183752, | |
| "eval_loss": 0.7894760370254517, | |
| "eval_runtime": 56.6018, | |
| "eval_samples_per_second": 1.767, | |
| "eval_steps_per_second": 0.23, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.38684719535783363, | |
| "grad_norm": 0.03189089894294739, | |
| "learning_rate": 0.0004842293906810036, | |
| "loss": 0.7802, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.38684719535783363, | |
| "eval_loss": 0.7783958315849304, | |
| "eval_runtime": 56.5799, | |
| "eval_samples_per_second": 1.767, | |
| "eval_steps_per_second": 0.23, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4835589941972921, | |
| "grad_norm": 0.03580320626497269, | |
| "learning_rate": 0.00046630824372759856, | |
| "loss": 0.7671, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.4835589941972921, | |
| "eval_loss": 0.7721498012542725, | |
| "eval_runtime": 57.9466, | |
| "eval_samples_per_second": 1.726, | |
| "eval_steps_per_second": 0.224, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5802707930367504, | |
| "grad_norm": 0.034203894436359406, | |
| "learning_rate": 0.00044838709677419355, | |
| "loss": 0.761, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5802707930367504, | |
| "eval_loss": 0.7687731385231018, | |
| "eval_runtime": 60.6867, | |
| "eval_samples_per_second": 1.648, | |
| "eval_steps_per_second": 0.214, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6769825918762089, | |
| "grad_norm": 0.03307868540287018, | |
| "learning_rate": 0.00043046594982078853, | |
| "loss": 0.7587, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.6769825918762089, | |
| "eval_loss": 0.766252338886261, | |
| "eval_runtime": 56.5991, | |
| "eval_samples_per_second": 1.767, | |
| "eval_steps_per_second": 0.23, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.7736943907156673, | |
| "grad_norm": 0.03772694244980812, | |
| "learning_rate": 0.0004125448028673835, | |
| "loss": 0.7529, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7736943907156673, | |
| "eval_loss": 0.7637045383453369, | |
| "eval_runtime": 56.5561, | |
| "eval_samples_per_second": 1.768, | |
| "eval_steps_per_second": 0.23, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8704061895551257, | |
| "grad_norm": 0.04153395816683769, | |
| "learning_rate": 0.00039462365591397855, | |
| "loss": 0.7562, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.8704061895551257, | |
| "eval_loss": 0.7616338133811951, | |
| "eval_runtime": 56.6718, | |
| "eval_samples_per_second": 1.765, | |
| "eval_steps_per_second": 0.229, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.9671179883945842, | |
| "grad_norm": 0.03528429567813873, | |
| "learning_rate": 0.00037670250896057353, | |
| "loss": 0.7507, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.9671179883945842, | |
| "eval_loss": 0.7601596117019653, | |
| "eval_runtime": 56.9725, | |
| "eval_samples_per_second": 1.755, | |
| "eval_steps_per_second": 0.228, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0638297872340425, | |
| "grad_norm": 0.03503500670194626, | |
| "learning_rate": 0.0003587813620071685, | |
| "loss": 0.7274, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.0638297872340425, | |
| "eval_loss": 0.7588610053062439, | |
| "eval_runtime": 60.6429, | |
| "eval_samples_per_second": 1.649, | |
| "eval_steps_per_second": 0.214, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.1605415860735009, | |
| "grad_norm": 0.0375722236931324, | |
| "learning_rate": 0.0003408602150537635, | |
| "loss": 0.7422, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.1605415860735009, | |
| "eval_loss": 0.7574188113212585, | |
| "eval_runtime": 56.7788, | |
| "eval_samples_per_second": 1.761, | |
| "eval_steps_per_second": 0.229, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.2572533849129595, | |
| "grad_norm": 0.04027143493294716, | |
| "learning_rate": 0.0003229390681003584, | |
| "loss": 0.735, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.2572533849129595, | |
| "eval_loss": 0.7570986747741699, | |
| "eval_runtime": 56.5753, | |
| "eval_samples_per_second": 1.768, | |
| "eval_steps_per_second": 0.23, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.3539651837524178, | |
| "grad_norm": 0.04008708521723747, | |
| "learning_rate": 0.0003050179211469534, | |
| "loss": 0.7367, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.3539651837524178, | |
| "eval_loss": 0.7555394172668457, | |
| "eval_runtime": 56.5444, | |
| "eval_samples_per_second": 1.769, | |
| "eval_steps_per_second": 0.23, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.4506769825918762, | |
| "grad_norm": 0.037061236798763275, | |
| "learning_rate": 0.0002870967741935484, | |
| "loss": 0.7471, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.4506769825918762, | |
| "eval_loss": 0.7549133896827698, | |
| "eval_runtime": 56.529, | |
| "eval_samples_per_second": 1.769, | |
| "eval_steps_per_second": 0.23, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.5473887814313345, | |
| "grad_norm": 0.04119856655597687, | |
| "learning_rate": 0.00026917562724014337, | |
| "loss": 0.7404, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.5473887814313345, | |
| "eval_loss": 0.7541146278381348, | |
| "eval_runtime": 60.5765, | |
| "eval_samples_per_second": 1.651, | |
| "eval_steps_per_second": 0.215, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.644100580270793, | |
| "grad_norm": 0.041316960006952286, | |
| "learning_rate": 0.00025125448028673835, | |
| "loss": 0.742, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.644100580270793, | |
| "eval_loss": 0.753339409828186, | |
| "eval_runtime": 56.7119, | |
| "eval_samples_per_second": 1.763, | |
| "eval_steps_per_second": 0.229, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.7408123791102514, | |
| "grad_norm": 0.04064641892910004, | |
| "learning_rate": 0.00023333333333333333, | |
| "loss": 0.7385, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.7408123791102514, | |
| "eval_loss": 0.7530195116996765, | |
| "eval_runtime": 56.5303, | |
| "eval_samples_per_second": 1.769, | |
| "eval_steps_per_second": 0.23, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.8375241779497098, | |
| "grad_norm": 0.03964291140437126, | |
| "learning_rate": 0.0002154121863799283, | |
| "loss": 0.7352, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.8375241779497098, | |
| "eval_loss": 0.7525290846824646, | |
| "eval_runtime": 56.698, | |
| "eval_samples_per_second": 1.764, | |
| "eval_steps_per_second": 0.229, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.9342359767891684, | |
| "grad_norm": 0.046565938740968704, | |
| "learning_rate": 0.0001974910394265233, | |
| "loss": 0.7323, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.9342359767891684, | |
| "eval_loss": 0.7516247630119324, | |
| "eval_runtime": 56.5619, | |
| "eval_samples_per_second": 1.768, | |
| "eval_steps_per_second": 0.23, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0309477756286265, | |
| "grad_norm": 0.04249171167612076, | |
| "learning_rate": 0.00017956989247311828, | |
| "loss": 0.7328, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.0309477756286265, | |
| "eval_loss": 0.7514955997467041, | |
| "eval_runtime": 60.5727, | |
| "eval_samples_per_second": 1.651, | |
| "eval_steps_per_second": 0.215, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.127659574468085, | |
| "grad_norm": 0.04245985299348831, | |
| "learning_rate": 0.00016164874551971326, | |
| "loss": 0.7264, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.127659574468085, | |
| "eval_loss": 0.7509506344795227, | |
| "eval_runtime": 56.8661, | |
| "eval_samples_per_second": 1.759, | |
| "eval_steps_per_second": 0.229, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.2243713733075436, | |
| "grad_norm": 0.04050629585981369, | |
| "learning_rate": 0.00014372759856630824, | |
| "loss": 0.704, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.2243713733075436, | |
| "eval_loss": 0.7505396008491516, | |
| "eval_runtime": 56.5726, | |
| "eval_samples_per_second": 1.768, | |
| "eval_steps_per_second": 0.23, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.3210831721470018, | |
| "grad_norm": 0.043631162494421005, | |
| "learning_rate": 0.00012580645161290322, | |
| "loss": 0.7242, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.3210831721470018, | |
| "eval_loss": 0.751037061214447, | |
| "eval_runtime": 56.6798, | |
| "eval_samples_per_second": 1.764, | |
| "eval_steps_per_second": 0.229, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.4177949709864603, | |
| "grad_norm": 0.04337254539132118, | |
| "learning_rate": 0.0001078853046594982, | |
| "loss": 0.7203, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.4177949709864603, | |
| "eval_loss": 0.7501779794692993, | |
| "eval_runtime": 56.6502, | |
| "eval_samples_per_second": 1.765, | |
| "eval_steps_per_second": 0.229, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.514506769825919, | |
| "grad_norm": 0.04406700283288956, | |
| "learning_rate": 8.996415770609319e-05, | |
| "loss": 0.7285, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.514506769825919, | |
| "eval_loss": 0.7498863339424133, | |
| "eval_runtime": 60.6054, | |
| "eval_samples_per_second": 1.65, | |
| "eval_steps_per_second": 0.215, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.611218568665377, | |
| "grad_norm": 0.04529803246259689, | |
| "learning_rate": 7.204301075268817e-05, | |
| "loss": 0.7192, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.611218568665377, | |
| "eval_loss": 0.7502144575119019, | |
| "eval_runtime": 57.5261, | |
| "eval_samples_per_second": 1.738, | |
| "eval_steps_per_second": 0.226, | |
| "step": 1350 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 1551, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.7515113721473024e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |