| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 100, | |
| "global_step": 248, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04032258064516129, | |
| "grad_norm": 1.8761260509490967, | |
| "learning_rate": 1.967741935483871e-05, | |
| "loss": 1.3894, | |
| "num_tokens": 583069.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.08064516129032258, | |
| "grad_norm": 1.5366759300231934, | |
| "learning_rate": 1.9274193548387097e-05, | |
| "loss": 1.2545, | |
| "num_tokens": 1166377.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.12096774193548387, | |
| "grad_norm": 1.5728304386138916, | |
| "learning_rate": 1.8870967741935487e-05, | |
| "loss": 1.2989, | |
| "num_tokens": 1764845.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.16129032258064516, | |
| "grad_norm": 1.7080962657928467, | |
| "learning_rate": 1.8467741935483873e-05, | |
| "loss": 1.2884, | |
| "num_tokens": 2359985.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.20161290322580644, | |
| "grad_norm": 1.4729899168014526, | |
| "learning_rate": 1.806451612903226e-05, | |
| "loss": 1.2504, | |
| "num_tokens": 2938303.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.24193548387096775, | |
| "grad_norm": 1.5811930894851685, | |
| "learning_rate": 1.7661290322580645e-05, | |
| "loss": 1.2157, | |
| "num_tokens": 3536824.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.28225806451612906, | |
| "grad_norm": 1.5144275426864624, | |
| "learning_rate": 1.7258064516129035e-05, | |
| "loss": 1.1586, | |
| "num_tokens": 4106428.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.3225806451612903, | |
| "grad_norm": 1.2803715467453003, | |
| "learning_rate": 1.685483870967742e-05, | |
| "loss": 1.2279, | |
| "num_tokens": 4713410.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3629032258064516, | |
| "grad_norm": 1.0752050876617432, | |
| "learning_rate": 1.6451612903225807e-05, | |
| "loss": 1.208, | |
| "num_tokens": 5302842.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.4032258064516129, | |
| "grad_norm": 0.8806188106536865, | |
| "learning_rate": 1.6048387096774193e-05, | |
| "loss": 1.1532, | |
| "num_tokens": 5883277.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.4435483870967742, | |
| "grad_norm": 0.7167242169380188, | |
| "learning_rate": 1.5645161290322583e-05, | |
| "loss": 1.16, | |
| "num_tokens": 6457162.0, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.4838709677419355, | |
| "grad_norm": 0.6805269122123718, | |
| "learning_rate": 1.5241935483870969e-05, | |
| "loss": 1.1472, | |
| "num_tokens": 7013145.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5241935483870968, | |
| "grad_norm": 0.7752795219421387, | |
| "learning_rate": 1.4838709677419357e-05, | |
| "loss": 1.1305, | |
| "num_tokens": 7621530.0, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.5645161290322581, | |
| "grad_norm": 0.6784509420394897, | |
| "learning_rate": 1.4435483870967743e-05, | |
| "loss": 1.1395, | |
| "num_tokens": 8194938.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6048387096774194, | |
| "grad_norm": 0.6889496445655823, | |
| "learning_rate": 1.4032258064516131e-05, | |
| "loss": 1.132, | |
| "num_tokens": 8800144.0, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.6451612903225806, | |
| "grad_norm": 0.7251303195953369, | |
| "learning_rate": 1.3629032258064517e-05, | |
| "loss": 1.1685, | |
| "num_tokens": 9408277.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.6854838709677419, | |
| "grad_norm": 0.6782833337783813, | |
| "learning_rate": 1.3225806451612903e-05, | |
| "loss": 1.1287, | |
| "num_tokens": 10000739.0, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.7258064516129032, | |
| "grad_norm": 0.6917558312416077, | |
| "learning_rate": 1.2822580645161291e-05, | |
| "loss": 1.1286, | |
| "num_tokens": 10586681.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.7661290322580645, | |
| "grad_norm": 0.693255603313446, | |
| "learning_rate": 1.2419354838709679e-05, | |
| "loss": 1.1328, | |
| "num_tokens": 11185064.0, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.8064516129032258, | |
| "grad_norm": 0.6752443313598633, | |
| "learning_rate": 1.2016129032258067e-05, | |
| "loss": 1.1341, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8064516129032258, | |
| "eval_loss": 1.121854543685913, | |
| "eval_num_tokens": 11776617.0, | |
| "eval_runtime": 0.7163, | |
| "eval_samples_per_second": 279.2, | |
| "eval_steps_per_second": 5.584, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8467741935483871, | |
| "grad_norm": 0.6200473308563232, | |
| "learning_rate": 1.1612903225806453e-05, | |
| "loss": 1.0986, | |
| "num_tokens": 12370086.0, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.8870967741935484, | |
| "grad_norm": 0.6284494996070862, | |
| "learning_rate": 1.1209677419354839e-05, | |
| "loss": 1.1591, | |
| "num_tokens": 12949436.0, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.9274193548387096, | |
| "grad_norm": 0.6932438611984253, | |
| "learning_rate": 1.0806451612903225e-05, | |
| "loss": 1.1478, | |
| "num_tokens": 13537898.0, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.967741935483871, | |
| "grad_norm": 0.6768374443054199, | |
| "learning_rate": 1.0403225806451613e-05, | |
| "loss": 1.1116, | |
| "num_tokens": 14126751.0, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0080645161290323, | |
| "grad_norm": 0.7337521910667419, | |
| "learning_rate": 1e-05, | |
| "loss": 1.0969, | |
| "num_tokens": 14734884.0, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.0483870967741935, | |
| "grad_norm": 0.7160388827323914, | |
| "learning_rate": 9.596774193548389e-06, | |
| "loss": 1.0158, | |
| "num_tokens": 15308122.0, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0887096774193548, | |
| "grad_norm": 0.6945542097091675, | |
| "learning_rate": 9.193548387096775e-06, | |
| "loss": 0.9843, | |
| "num_tokens": 15893652.0, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.129032258064516, | |
| "grad_norm": 0.6891147494316101, | |
| "learning_rate": 8.790322580645163e-06, | |
| "loss": 1.0347, | |
| "num_tokens": 16479176.0, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.1693548387096775, | |
| "grad_norm": 0.6116432547569275, | |
| "learning_rate": 8.387096774193549e-06, | |
| "loss": 0.9615, | |
| "num_tokens": 17068227.0, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.2096774193548387, | |
| "grad_norm": 0.6009991765022278, | |
| "learning_rate": 7.983870967741935e-06, | |
| "loss": 0.9883, | |
| "num_tokens": 17648475.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.6140655875205994, | |
| "learning_rate": 7.580645161290323e-06, | |
| "loss": 0.9833, | |
| "num_tokens": 18252289.0, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 1.2903225806451613, | |
| "grad_norm": 0.5997236967086792, | |
| "learning_rate": 7.177419354838711e-06, | |
| "loss": 0.9659, | |
| "num_tokens": 18842881.0, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.3306451612903225, | |
| "grad_norm": 0.583691418170929, | |
| "learning_rate": 6.774193548387097e-06, | |
| "loss": 1.0074, | |
| "num_tokens": 19423540.0, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 1.370967741935484, | |
| "grad_norm": 0.617231011390686, | |
| "learning_rate": 6.370967741935485e-06, | |
| "loss": 1.0354, | |
| "num_tokens": 20017275.0, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.4112903225806452, | |
| "grad_norm": 0.6525187492370605, | |
| "learning_rate": 5.967741935483872e-06, | |
| "loss": 1.0141, | |
| "num_tokens": 20592466.0, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.4516129032258065, | |
| "grad_norm": 0.6343702673912048, | |
| "learning_rate": 5.564516129032258e-06, | |
| "loss": 1.0358, | |
| "num_tokens": 21170579.0, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.4919354838709677, | |
| "grad_norm": 0.621397078037262, | |
| "learning_rate": 5.161290322580646e-06, | |
| "loss": 0.982, | |
| "num_tokens": 21779118.0, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.532258064516129, | |
| "grad_norm": 0.6357585191726685, | |
| "learning_rate": 4.758064516129033e-06, | |
| "loss": 0.9599, | |
| "num_tokens": 22337296.0, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.5725806451612905, | |
| "grad_norm": 0.6274889707565308, | |
| "learning_rate": 4.35483870967742e-06, | |
| "loss": 1.0216, | |
| "num_tokens": 22933384.0, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 1.6129032258064515, | |
| "grad_norm": 0.6913103461265564, | |
| "learning_rate": 3.951612903225807e-06, | |
| "loss": 0.9954, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6129032258064515, | |
| "eval_loss": 1.1182941198349, | |
| "eval_num_tokens": 23494529.0, | |
| "eval_runtime": 0.6682, | |
| "eval_samples_per_second": 299.331, | |
| "eval_steps_per_second": 5.987, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.653225806451613, | |
| "grad_norm": 0.6222682595252991, | |
| "learning_rate": 3.548387096774194e-06, | |
| "loss": 1.0195, | |
| "num_tokens": 24073440.0, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 1.6935483870967742, | |
| "grad_norm": 0.6264064311981201, | |
| "learning_rate": 3.145161290322581e-06, | |
| "loss": 1.0458, | |
| "num_tokens": 24679112.0, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.7338709677419355, | |
| "grad_norm": 0.629045844078064, | |
| "learning_rate": 2.7419354838709676e-06, | |
| "loss": 1.028, | |
| "num_tokens": 25286383.0, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.7741935483870968, | |
| "grad_norm": 0.575624406337738, | |
| "learning_rate": 2.338709677419355e-06, | |
| "loss": 0.9633, | |
| "num_tokens": 25917151.0, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.814516129032258, | |
| "grad_norm": 0.5966957211494446, | |
| "learning_rate": 1.935483870967742e-06, | |
| "loss": 1.0002, | |
| "num_tokens": 26525780.0, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.8548387096774195, | |
| "grad_norm": 0.6634185314178467, | |
| "learning_rate": 1.5322580645161292e-06, | |
| "loss": 0.998, | |
| "num_tokens": 27096606.0, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.8951612903225805, | |
| "grad_norm": 0.5811392068862915, | |
| "learning_rate": 1.1290322580645162e-06, | |
| "loss": 1.0284, | |
| "num_tokens": 27687166.0, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.935483870967742, | |
| "grad_norm": 0.6014962792396545, | |
| "learning_rate": 7.258064516129033e-07, | |
| "loss": 0.9879, | |
| "num_tokens": 28284352.0, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.9758064516129032, | |
| "grad_norm": 0.5961019396781921, | |
| "learning_rate": 3.2258064516129035e-07, | |
| "loss": 0.9683, | |
| "num_tokens": 28878645.0, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "num_tokens": 29227434.0, | |
| "step": 248, | |
| "total_flos": 4.8695749825160806e+17, | |
| "train_loss": 1.0886195955737945, | |
| "train_runtime": 315.8591, | |
| "train_samples_per_second": 100.083, | |
| "train_steps_per_second": 0.785 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 248, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "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": 4.8695749825160806e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |