| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.0, | |
| "eval_steps": 500, | |
| "global_step": 4578, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.764525993883792, | |
| "grad_norm": 2.860719919204712, | |
| "learning_rate": 1.781564001747488e-05, | |
| "loss": 0.2481, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9536874012447196, | |
| "eval_f1": 0.8924138535986316, | |
| "eval_loss": 0.13802066445350647, | |
| "eval_precision": 0.8979934094386254, | |
| "eval_recall": 0.8869032051840874, | |
| "eval_runtime": 25.0596, | |
| "eval_samples_per_second": 212.693, | |
| "eval_steps_per_second": 26.617, | |
| "step": 654 | |
| }, | |
| { | |
| "epoch": 1.529051987767584, | |
| "grad_norm": 2.250061273574829, | |
| "learning_rate": 1.563128003494976e-05, | |
| "loss": 0.0943, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9569120634613524, | |
| "eval_f1": 0.9029714891848114, | |
| "eval_loss": 0.1516450196504593, | |
| "eval_precision": 0.891004243281471, | |
| "eval_recall": 0.9152645802458373, | |
| "eval_runtime": 25.372, | |
| "eval_samples_per_second": 210.074, | |
| "eval_steps_per_second": 26.289, | |
| "step": 1308 | |
| }, | |
| { | |
| "epoch": 2.293577981651376, | |
| "grad_norm": 0.9499523043632507, | |
| "learning_rate": 1.344692005242464e-05, | |
| "loss": 0.0595, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9584212053787365, | |
| "eval_f1": 0.9057778035083657, | |
| "eval_loss": 0.17137625813484192, | |
| "eval_precision": 0.9023039880042677, | |
| "eval_recall": 0.909278470345509, | |
| "eval_runtime": 24.0746, | |
| "eval_samples_per_second": 221.395, | |
| "eval_steps_per_second": 27.706, | |
| "step": 1962 | |
| }, | |
| { | |
| "epoch": 3.058103975535168, | |
| "grad_norm": 0.29300621151924133, | |
| "learning_rate": 1.126256006989952e-05, | |
| "loss": 0.0405, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.8226299694189603, | |
| "grad_norm": 0.09997886419296265, | |
| "learning_rate": 9.0782000873744e-06, | |
| "loss": 0.0218, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9578988100996421, | |
| "eval_f1": 0.9073512721421696, | |
| "eval_loss": 0.20640073716640472, | |
| "eval_precision": 0.8972836278911178, | |
| "eval_recall": 0.9176474006915991, | |
| "eval_runtime": 23.267, | |
| "eval_samples_per_second": 229.08, | |
| "eval_steps_per_second": 28.667, | |
| "step": 2616 | |
| }, | |
| { | |
| "epoch": 4.587155963302752, | |
| "grad_norm": 1.0946885347366333, | |
| "learning_rate": 6.89384010484928e-06, | |
| "loss": 0.0121, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9587372222759666, | |
| "eval_f1": 0.9086317256194029, | |
| "eval_loss": 0.22833754122257233, | |
| "eval_precision": 0.9076963229323745, | |
| "eval_recall": 0.9095690582047482, | |
| "eval_runtime": 23.0423, | |
| "eval_samples_per_second": 231.314, | |
| "eval_steps_per_second": 28.947, | |
| "step": 3270 | |
| }, | |
| { | |
| "epoch": 5.351681957186544, | |
| "grad_norm": 0.06167488545179367, | |
| "learning_rate": 4.70948012232416e-06, | |
| "loss": 0.0106, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9577440263132437, | |
| "eval_f1": 0.9077642656688494, | |
| "eval_loss": 0.24395699799060822, | |
| "eval_precision": 0.8992358576642335, | |
| "eval_recall": 0.9164559904687182, | |
| "eval_runtime": 29.2986, | |
| "eval_samples_per_second": 181.92, | |
| "eval_steps_per_second": 22.766, | |
| "step": 3924 | |
| }, | |
| { | |
| "epoch": 6.116207951070336, | |
| "grad_norm": 0.0039491127245128155, | |
| "learning_rate": 2.5251201397990393e-06, | |
| "loss": 0.0067, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.8807339449541285, | |
| "grad_norm": 0.015259736217558384, | |
| "learning_rate": 3.407601572739188e-07, | |
| "loss": 0.005, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9576085905001451, | |
| "eval_f1": 0.9074234296591107, | |
| "eval_loss": 0.2515053153038025, | |
| "eval_precision": 0.9006440532417347, | |
| "eval_recall": 0.9143056403103479, | |
| "eval_runtime": 24.477, | |
| "eval_samples_per_second": 217.756, | |
| "eval_steps_per_second": 27.25, | |
| "step": 4578 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4578, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "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": 1242291550958640.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |