| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.847457627118644, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00847457627118644, | |
| "eval_loss": 1.2892621755599976, | |
| "eval_runtime": 23.2041, | |
| "eval_samples_per_second": 18.1, | |
| "eval_steps_per_second": 2.284, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0423728813559322, | |
| "grad_norm": 0.8503715395927429, | |
| "learning_rate": 3.75e-05, | |
| "loss": 1.2136, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.07627118644067797, | |
| "eval_loss": 1.1160207986831665, | |
| "eval_runtime": 22.2287, | |
| "eval_samples_per_second": 18.894, | |
| "eval_steps_per_second": 2.384, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0847457627118644, | |
| "grad_norm": 0.8837136626243591, | |
| "learning_rate": 7.5e-05, | |
| "loss": 1.1313, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1271186440677966, | |
| "grad_norm": 0.7536290287971497, | |
| "learning_rate": 0.0001125, | |
| "loss": 1.0435, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.15254237288135594, | |
| "eval_loss": 0.9707416296005249, | |
| "eval_runtime": 22.2197, | |
| "eval_samples_per_second": 18.902, | |
| "eval_steps_per_second": 2.385, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.1694915254237288, | |
| "grad_norm": 0.8838627338409424, | |
| "learning_rate": 0.00015, | |
| "loss": 0.9562, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.211864406779661, | |
| "grad_norm": 0.6448015570640564, | |
| "learning_rate": 0.00014855889603024227, | |
| "loss": 0.884, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.2288135593220339, | |
| "eval_loss": 0.9067414999008179, | |
| "eval_runtime": 22.2329, | |
| "eval_samples_per_second": 18.891, | |
| "eval_steps_per_second": 2.384, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.2542372881355932, | |
| "grad_norm": 0.6635559797286987, | |
| "learning_rate": 0.0001442909649383465, | |
| "loss": 0.9048, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2966101694915254, | |
| "grad_norm": 0.6447595357894897, | |
| "learning_rate": 0.0001373602209226909, | |
| "loss": 0.8876, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.3050847457627119, | |
| "eval_loss": 0.8686102032661438, | |
| "eval_runtime": 22.2429, | |
| "eval_samples_per_second": 18.882, | |
| "eval_steps_per_second": 2.383, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.3389830508474576, | |
| "grad_norm": 0.5714731216430664, | |
| "learning_rate": 0.00012803300858899104, | |
| "loss": 0.8314, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3813559322033898, | |
| "grad_norm": 0.5815892219543457, | |
| "learning_rate": 0.00011666776747647015, | |
| "loss": 0.7683, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.3813559322033898, | |
| "eval_loss": 0.8386306166648865, | |
| "eval_runtime": 22.2395, | |
| "eval_samples_per_second": 18.885, | |
| "eval_steps_per_second": 2.383, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.423728813559322, | |
| "grad_norm": 0.5933341979980469, | |
| "learning_rate": 0.00010370125742738173, | |
| "loss": 0.792, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.4576271186440678, | |
| "eval_loss": 0.81903076171875, | |
| "eval_runtime": 22.2328, | |
| "eval_samples_per_second": 18.891, | |
| "eval_steps_per_second": 2.384, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.4661016949152542, | |
| "grad_norm": 0.5448825359344482, | |
| "learning_rate": 8.963177415120962e-05, | |
| "loss": 0.7919, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.5084745762711864, | |
| "grad_norm": 0.5443424582481384, | |
| "learning_rate": 7.5e-05, | |
| "loss": 0.7779, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5338983050847458, | |
| "eval_loss": 0.799988329410553, | |
| "eval_runtime": 22.2537, | |
| "eval_samples_per_second": 18.873, | |
| "eval_steps_per_second": 2.382, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.5508474576271186, | |
| "grad_norm": 0.5052010416984558, | |
| "learning_rate": 6.036822584879038e-05, | |
| "loss": 0.7611, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.5932203389830508, | |
| "grad_norm": 0.5269892811775208, | |
| "learning_rate": 4.6298742572618266e-05, | |
| "loss": 0.7769, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6101694915254238, | |
| "eval_loss": 0.7839500904083252, | |
| "eval_runtime": 22.2533, | |
| "eval_samples_per_second": 18.874, | |
| "eval_steps_per_second": 2.382, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.635593220338983, | |
| "grad_norm": 0.5438559055328369, | |
| "learning_rate": 3.333223252352985e-05, | |
| "loss": 0.7603, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.6779661016949152, | |
| "grad_norm": 0.5549936294555664, | |
| "learning_rate": 2.1966991411008938e-05, | |
| "loss": 0.7573, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.6864406779661016, | |
| "eval_loss": 0.7745689153671265, | |
| "eval_runtime": 22.2322, | |
| "eval_samples_per_second": 18.891, | |
| "eval_steps_per_second": 2.384, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.7203389830508474, | |
| "grad_norm": 0.5136195421218872, | |
| "learning_rate": 1.2639779077309098e-05, | |
| "loss": 0.77, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.7627118644067796, | |
| "grad_norm": 0.5830088257789612, | |
| "learning_rate": 5.709035061653494e-06, | |
| "loss": 0.7518, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.7627118644067796, | |
| "eval_loss": 0.7687093019485474, | |
| "eval_runtime": 22.2345, | |
| "eval_samples_per_second": 18.89, | |
| "eval_steps_per_second": 2.384, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8050847457627118, | |
| "grad_norm": 0.5288379192352295, | |
| "learning_rate": 1.4411039697577175e-06, | |
| "loss": 0.7473, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.8389830508474576, | |
| "eval_loss": 0.7672185897827148, | |
| "eval_runtime": 22.2757, | |
| "eval_samples_per_second": 18.855, | |
| "eval_steps_per_second": 2.379, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.847457627118644, | |
| "grad_norm": 0.5222312211990356, | |
| "learning_rate": 0.0, | |
| "loss": 0.7123, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 1.549997702161367e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |