{ "best_metric": null, "best_model_checkpoint": null, "epoch": 10.0, "eval_steps": 500, "global_step": 590, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.3389830508474576, "grad_norm": 10.4375, "learning_rate": 4.8305084745762714e-05, "loss": 1.4141, "step": 20 }, { "epoch": 0.6779661016949152, "grad_norm": 16.125, "learning_rate": 4.6610169491525425e-05, "loss": 0.7498, "step": 40 }, { "epoch": 1.0169491525423728, "grad_norm": 2.59375, "learning_rate": 4.491525423728814e-05, "loss": 0.2804, "step": 60 }, { "epoch": 1.3559322033898304, "grad_norm": 2.15625, "learning_rate": 4.3220338983050854e-05, "loss": 0.106, "step": 80 }, { "epoch": 1.694915254237288, "grad_norm": 2.203125, "learning_rate": 4.152542372881356e-05, "loss": 0.1093, "step": 100 }, { "epoch": 2.0338983050847457, "grad_norm": 1.7734375, "learning_rate": 3.983050847457627e-05, "loss": 0.1021, "step": 120 }, { "epoch": 2.3728813559322033, "grad_norm": 1.640625, "learning_rate": 3.813559322033898e-05, "loss": 0.0724, "step": 140 }, { "epoch": 2.711864406779661, "grad_norm": 1.8515625, "learning_rate": 3.644067796610169e-05, "loss": 0.0701, "step": 160 }, { "epoch": 3.0508474576271185, "grad_norm": 1.4921875, "learning_rate": 3.474576271186441e-05, "loss": 0.071, "step": 180 }, { "epoch": 3.389830508474576, "grad_norm": 1.4296875, "learning_rate": 3.305084745762712e-05, "loss": 0.0552, "step": 200 }, { "epoch": 3.7288135593220337, "grad_norm": 1.6953125, "learning_rate": 3.135593220338983e-05, "loss": 0.0517, "step": 220 }, { "epoch": 4.067796610169491, "grad_norm": 1.71875, "learning_rate": 2.9661016949152544e-05, "loss": 0.0553, "step": 240 }, { "epoch": 4.406779661016949, "grad_norm": 1.2265625, "learning_rate": 2.7966101694915255e-05, "loss": 0.0422, "step": 260 }, { "epoch": 4.745762711864407, "grad_norm": 1.2734375, "learning_rate": 2.627118644067797e-05, "loss": 0.0459, "step": 280 }, { "epoch": 5.084745762711864, "grad_norm": 1.0859375, "learning_rate": 2.457627118644068e-05, "loss": 0.0431, "step": 300 }, { "epoch": 5.423728813559322, "grad_norm": 1.4609375, "learning_rate": 2.2881355932203392e-05, "loss": 0.0366, "step": 320 }, { "epoch": 5.762711864406779, "grad_norm": 0.97265625, "learning_rate": 2.1186440677966103e-05, "loss": 0.0373, "step": 340 }, { "epoch": 6.101694915254237, "grad_norm": 0.97265625, "learning_rate": 1.9491525423728814e-05, "loss": 0.0403, "step": 360 }, { "epoch": 6.440677966101695, "grad_norm": 1.21875, "learning_rate": 1.7796610169491526e-05, "loss": 0.0334, "step": 380 }, { "epoch": 6.779661016949152, "grad_norm": 1.4140625, "learning_rate": 1.6101694915254237e-05, "loss": 0.0364, "step": 400 }, { "epoch": 7.11864406779661, "grad_norm": 1.3203125, "learning_rate": 1.440677966101695e-05, "loss": 0.0349, "step": 420 }, { "epoch": 7.4576271186440675, "grad_norm": 1.1640625, "learning_rate": 1.2711864406779661e-05, "loss": 0.0289, "step": 440 }, { "epoch": 7.796610169491525, "grad_norm": 1.4453125, "learning_rate": 1.1016949152542374e-05, "loss": 0.0311, "step": 460 }, { "epoch": 8.135593220338983, "grad_norm": 0.9140625, "learning_rate": 9.322033898305085e-06, "loss": 0.0328, "step": 480 }, { "epoch": 8.474576271186441, "grad_norm": 1.3984375, "learning_rate": 7.627118644067798e-06, "loss": 0.0331, "step": 500 }, { "epoch": 8.813559322033898, "grad_norm": 1.203125, "learning_rate": 5.932203389830509e-06, "loss": 0.0315, "step": 520 }, { "epoch": 9.152542372881356, "grad_norm": 1.5546875, "learning_rate": 4.23728813559322e-06, "loss": 0.0302, "step": 540 }, { "epoch": 9.491525423728813, "grad_norm": 0.90234375, "learning_rate": 2.5423728813559323e-06, "loss": 0.0313, "step": 560 }, { "epoch": 9.830508474576272, "grad_norm": 1.2421875, "learning_rate": 8.474576271186441e-07, "loss": 0.0307, "step": 580 } ], "logging_steps": 20, "max_steps": 590, "num_input_tokens_seen": 0, "num_train_epochs": 10, "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": 2.4341956558848e+18, "train_batch_size": 64, "trial_name": null, "trial_params": null }