| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.6315789473684212, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02631578947368421, | |
| "eval_loss": 1.6184024810791016, | |
| "eval_runtime": 4.386, | |
| "eval_samples_per_second": 14.592, | |
| "eval_steps_per_second": 1.824, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07894736842105263, | |
| "grad_norm": 25.33377456665039, | |
| "learning_rate": 3e-05, | |
| "loss": 6.2534, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.15789473684210525, | |
| "grad_norm": 19.7912540435791, | |
| "learning_rate": 6e-05, | |
| "loss": 6.0633, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.23684210526315788, | |
| "grad_norm": 17.20012855529785, | |
| "learning_rate": 9e-05, | |
| "loss": 5.0168, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.23684210526315788, | |
| "eval_loss": 1.0376337766647339, | |
| "eval_runtime": 4.4723, | |
| "eval_samples_per_second": 14.31, | |
| "eval_steps_per_second": 1.789, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3157894736842105, | |
| "grad_norm": 11.844976425170898, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 4.0152, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.39473684210526316, | |
| "grad_norm": 10.39311695098877, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 3.859, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.47368421052631576, | |
| "grad_norm": 9.025694847106934, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 3.9375, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.47368421052631576, | |
| "eval_loss": 0.8340747356414795, | |
| "eval_runtime": 4.4933, | |
| "eval_samples_per_second": 14.244, | |
| "eval_steps_per_second": 1.78, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.5526315789473685, | |
| "grad_norm": 8.11834716796875, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 3.2265, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.631578947368421, | |
| "grad_norm": 10.354762077331543, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 3.3718, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.7105263157894737, | |
| "grad_norm": 8.096390724182129, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 3.5513, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.7105263157894737, | |
| "eval_loss": 0.8040823340415955, | |
| "eval_runtime": 4.4907, | |
| "eval_samples_per_second": 14.252, | |
| "eval_steps_per_second": 1.781, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.7894736842105263, | |
| "grad_norm": 8.748348236083984, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 3.7277, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.868421052631579, | |
| "grad_norm": 9.880663871765137, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 3.2257, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.9473684210526315, | |
| "grad_norm": 8.828300476074219, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 3.4785, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.9473684210526315, | |
| "eval_loss": 0.7874240279197693, | |
| "eval_runtime": 4.495, | |
| "eval_samples_per_second": 14.238, | |
| "eval_steps_per_second": 1.78, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.0263157894736843, | |
| "grad_norm": 6.3108296394348145, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 3.1481, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.1052631578947367, | |
| "grad_norm": 8.543023109436035, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 2.5156, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.1842105263157894, | |
| "grad_norm": 6.542039394378662, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 2.4297, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.1842105263157894, | |
| "eval_loss": 0.7846142649650574, | |
| "eval_runtime": 4.4938, | |
| "eval_samples_per_second": 14.242, | |
| "eval_steps_per_second": 1.78, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.263157894736842, | |
| "grad_norm": 7.43921422958374, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 2.4722, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.3421052631578947, | |
| "grad_norm": 8.37086296081543, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 2.6133, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.4210526315789473, | |
| "grad_norm": 8.553829193115234, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 2.6193, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.4210526315789473, | |
| "eval_loss": 0.7718786597251892, | |
| "eval_runtime": 4.4967, | |
| "eval_samples_per_second": 14.233, | |
| "eval_steps_per_second": 1.779, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 6.831599712371826, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 2.3717, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.5789473684210527, | |
| "grad_norm": 8.077478408813477, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 2.124, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.6578947368421053, | |
| "grad_norm": 7.879061222076416, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 2.1771, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.6578947368421053, | |
| "eval_loss": 0.7753761410713196, | |
| "eval_runtime": 4.4951, | |
| "eval_samples_per_second": 14.238, | |
| "eval_steps_per_second": 1.78, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.736842105263158, | |
| "grad_norm": 9.225750923156738, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 2.2392, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.8157894736842106, | |
| "grad_norm": 8.263725280761719, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 2.1842, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.8947368421052633, | |
| "grad_norm": 8.750575065612793, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 2.3606, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.8947368421052633, | |
| "eval_loss": 0.7717390060424805, | |
| "eval_runtime": 4.4993, | |
| "eval_samples_per_second": 14.225, | |
| "eval_steps_per_second": 1.778, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.973684210526316, | |
| "grad_norm": 7.726223468780518, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 1.9729, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.0526315789473686, | |
| "grad_norm": 6.892090320587158, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 1.7884, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.1315789473684212, | |
| "grad_norm": 7.418595790863037, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 1.508, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.1315789473684212, | |
| "eval_loss": 0.7756311297416687, | |
| "eval_runtime": 4.4917, | |
| "eval_samples_per_second": 14.249, | |
| "eval_steps_per_second": 1.781, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.2105263157894735, | |
| "grad_norm": 7.621229648590088, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 1.5441, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.2894736842105265, | |
| "grad_norm": 6.870114803314209, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 1.5711, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 2.3684210526315788, | |
| "grad_norm": 7.845399379730225, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 1.504, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.3684210526315788, | |
| "eval_loss": 0.7961782217025757, | |
| "eval_runtime": 4.497, | |
| "eval_samples_per_second": 14.232, | |
| "eval_steps_per_second": 1.779, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.4473684210526314, | |
| "grad_norm": 7.044579982757568, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 1.5221, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 2.526315789473684, | |
| "grad_norm": 7.962366580963135, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 1.4521, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.6052631578947367, | |
| "grad_norm": 7.687761306762695, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 1.473, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 2.6052631578947367, | |
| "eval_loss": 0.8015087842941284, | |
| "eval_runtime": 4.5008, | |
| "eval_samples_per_second": 14.22, | |
| "eval_steps_per_second": 1.777, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.4001082178155315e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |