| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.14705882352941177, | |
| "eval_steps": 4, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004901960784313725, | |
| "grad_norm": 1.948913812637329, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 1.0601, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004901960784313725, | |
| "eval_loss": 0.9626200199127197, | |
| "eval_runtime": 11.2758, | |
| "eval_samples_per_second": 7.627, | |
| "eval_steps_per_second": 1.951, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00980392156862745, | |
| "grad_norm": 2.0441207885742188, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.9237, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.014705882352941176, | |
| "grad_norm": 1.0346183776855469, | |
| "learning_rate": 6e-06, | |
| "loss": 1.7715, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0196078431372549, | |
| "grad_norm": 1.7915290594100952, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.9189, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0196078431372549, | |
| "eval_loss": 0.9602966904640198, | |
| "eval_runtime": 10.5025, | |
| "eval_samples_per_second": 8.189, | |
| "eval_steps_per_second": 2.095, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.024509803921568627, | |
| "grad_norm": 1.6641079187393188, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7885, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.029411764705882353, | |
| "grad_norm": 1.4073126316070557, | |
| "learning_rate": 9.960573506572391e-06, | |
| "loss": 0.6591, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.03431372549019608, | |
| "grad_norm": 1.7350611686706543, | |
| "learning_rate": 9.842915805643156e-06, | |
| "loss": 0.7968, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0392156862745098, | |
| "grad_norm": 1.7413434982299805, | |
| "learning_rate": 9.648882429441258e-06, | |
| "loss": 0.8927, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0392156862745098, | |
| "eval_loss": 0.934045672416687, | |
| "eval_runtime": 10.5931, | |
| "eval_samples_per_second": 8.119, | |
| "eval_steps_per_second": 2.077, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.04411764705882353, | |
| "grad_norm": 1.2970942258834839, | |
| "learning_rate": 9.381533400219319e-06, | |
| "loss": 0.6153, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.049019607843137254, | |
| "grad_norm": 2.589458465576172, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 1.0173, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05392156862745098, | |
| "grad_norm": 1.39173424243927, | |
| "learning_rate": 8.644843137107058e-06, | |
| "loss": 0.7402, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.058823529411764705, | |
| "grad_norm": 1.885929822921753, | |
| "learning_rate": 8.18711994874345e-06, | |
| "loss": 0.6374, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.058823529411764705, | |
| "eval_loss": 0.9021902680397034, | |
| "eval_runtime": 10.6144, | |
| "eval_samples_per_second": 8.102, | |
| "eval_steps_per_second": 2.073, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.06372549019607843, | |
| "grad_norm": 1.750763177871704, | |
| "learning_rate": 7.679133974894984e-06, | |
| "loss": 0.8322, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.06862745098039216, | |
| "grad_norm": 1.578403353691101, | |
| "learning_rate": 7.128896457825364e-06, | |
| "loss": 0.8538, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.07352941176470588, | |
| "grad_norm": 1.8658047914505005, | |
| "learning_rate": 6.545084971874738e-06, | |
| "loss": 0.7806, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0784313725490196, | |
| "grad_norm": 2.268649101257324, | |
| "learning_rate": 5.936906572928625e-06, | |
| "loss": 0.9588, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0784313725490196, | |
| "eval_loss": 0.8679822087287903, | |
| "eval_runtime": 10.5908, | |
| "eval_samples_per_second": 8.12, | |
| "eval_steps_per_second": 2.077, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.08333333333333333, | |
| "grad_norm": 0.655454695224762, | |
| "learning_rate": 5.3139525976465675e-06, | |
| "loss": 0.7884, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.08823529411764706, | |
| "grad_norm": 2.9148428440093994, | |
| "learning_rate": 4.686047402353433e-06, | |
| "loss": 0.9067, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.09313725490196079, | |
| "grad_norm": 0.9642003774642944, | |
| "learning_rate": 4.063093427071376e-06, | |
| "loss": 0.636, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.09803921568627451, | |
| "grad_norm": 1.121366262435913, | |
| "learning_rate": 3.4549150281252635e-06, | |
| "loss": 0.7324, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09803921568627451, | |
| "eval_loss": 0.8456867933273315, | |
| "eval_runtime": 10.6004, | |
| "eval_samples_per_second": 8.113, | |
| "eval_steps_per_second": 2.075, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.10294117647058823, | |
| "grad_norm": 2.0282931327819824, | |
| "learning_rate": 2.871103542174637e-06, | |
| "loss": 0.8578, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.10784313725490197, | |
| "grad_norm": 0.8347979784011841, | |
| "learning_rate": 2.320866025105016e-06, | |
| "loss": 0.775, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.11274509803921569, | |
| "grad_norm": 1.1945385932922363, | |
| "learning_rate": 1.8128800512565514e-06, | |
| "loss": 0.6854, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.11764705882352941, | |
| "grad_norm": 1.330310583114624, | |
| "learning_rate": 1.3551568628929434e-06, | |
| "loss": 0.8785, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.11764705882352941, | |
| "eval_loss": 0.8344882726669312, | |
| "eval_runtime": 10.6019, | |
| "eval_samples_per_second": 8.112, | |
| "eval_steps_per_second": 2.075, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.12254901960784313, | |
| "grad_norm": 1.4803705215454102, | |
| "learning_rate": 9.549150281252633e-07, | |
| "loss": 0.6359, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.12745098039215685, | |
| "grad_norm": 1.536838173866272, | |
| "learning_rate": 6.184665997806832e-07, | |
| "loss": 0.6286, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.1323529411764706, | |
| "grad_norm": 1.0436162948608398, | |
| "learning_rate": 3.511175705587433e-07, | |
| "loss": 0.7679, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.13725490196078433, | |
| "grad_norm": 2.5319721698760986, | |
| "learning_rate": 1.5708419435684463e-07, | |
| "loss": 1.1476, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.13725490196078433, | |
| "eval_loss": 0.8303804397583008, | |
| "eval_runtime": 10.618, | |
| "eval_samples_per_second": 8.099, | |
| "eval_steps_per_second": 2.072, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.14215686274509803, | |
| "grad_norm": 1.8032675981521606, | |
| "learning_rate": 3.9426493427611177e-08, | |
| "loss": 0.8761, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.14705882352941177, | |
| "grad_norm": 2.128256320953369, | |
| "learning_rate": 0.0, | |
| "loss": 1.0381, | |
| "step": 30 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 20, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.041223234093056e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |