my_saved_model/checkpoint-5688/trainer_state.json

{
  "best_metric": 0.4109877347946167,
  "best_model_checkpoint": "./my_saved_model/checkpoint-5688",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 5688,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.7911392405063291,
      "grad_norm": 2.325883388519287,
      "learning_rate": 4.604430379746836e-05,
      "loss": 1.123,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.688195991091314,
      "eval_loss": 0.6270885467529297,
      "eval_runtime": 70.865,
      "eval_samples_per_second": 63.36,
      "eval_steps_per_second": 1.002,
      "step": 632
    },
    {
      "epoch": 1.5822784810126582,
      "grad_norm": 1.8189690113067627,
      "learning_rate": 4.208860759493671e-05,
      "loss": 0.6186,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8383073496659242,
      "eval_loss": 0.5550462007522583,
      "eval_runtime": 70.8726,
      "eval_samples_per_second": 63.353,
      "eval_steps_per_second": 1.002,
      "step": 1264
    },
    {
      "epoch": 2.3734177215189876,
      "grad_norm": 1.0061662197113037,
      "learning_rate": 3.813291139240506e-05,
      "loss": 0.5672,
      "step": 1500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8472160356347439,
      "eval_loss": 0.5077998042106628,
      "eval_runtime": 71.0525,
      "eval_samples_per_second": 63.193,
      "eval_steps_per_second": 0.999,
      "step": 1896
    },
    {
      "epoch": 3.1645569620253164,
      "grad_norm": 0.8065588474273682,
      "learning_rate": 3.4177215189873416e-05,
      "loss": 0.5307,
      "step": 2000
    },
    {
      "epoch": 3.9556962025316453,
      "grad_norm": 1.4700500965118408,
      "learning_rate": 3.0221518987341774e-05,
      "loss": 0.5023,
      "step": 2500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8688195991091314,
      "eval_loss": 0.47517454624176025,
      "eval_runtime": 70.885,
      "eval_samples_per_second": 63.342,
      "eval_steps_per_second": 1.002,
      "step": 2528
    },
    {
      "epoch": 4.746835443037975,
      "grad_norm": 0.5362122654914856,
      "learning_rate": 2.626582278481013e-05,
      "loss": 0.4817,
      "step": 3000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8748329621380846,
      "eval_loss": 0.45159268379211426,
      "eval_runtime": 71.0334,
      "eval_samples_per_second": 63.21,
      "eval_steps_per_second": 1.0,
      "step": 3160
    },
    {
      "epoch": 5.537974683544304,
      "grad_norm": 0.7847274541854858,
      "learning_rate": 2.231012658227848e-05,
      "loss": 0.4665,
      "step": 3500
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8792873051224944,
      "eval_loss": 0.4351290166378021,
      "eval_runtime": 70.9077,
      "eval_samples_per_second": 63.322,
      "eval_steps_per_second": 1.001,
      "step": 3792
    },
    {
      "epoch": 6.329113924050633,
      "grad_norm": 0.6061320304870605,
      "learning_rate": 1.8354430379746836e-05,
      "loss": 0.4567,
      "step": 4000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8797327394209354,
      "eval_loss": 0.422820121049881,
      "eval_runtime": 71.2032,
      "eval_samples_per_second": 63.059,
      "eval_steps_per_second": 0.997,
      "step": 4424
    },
    {
      "epoch": 7.120253164556962,
      "grad_norm": 1.0778601169586182,
      "learning_rate": 1.439873417721519e-05,
      "loss": 0.4469,
      "step": 4500
    },
    {
      "epoch": 7.911392405063291,
      "grad_norm": 1.6876882314682007,
      "learning_rate": 1.0443037974683544e-05,
      "loss": 0.4404,
      "step": 5000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8832962138084632,
      "eval_loss": 0.4156254827976227,
      "eval_runtime": 70.8877,
      "eval_samples_per_second": 63.34,
      "eval_steps_per_second": 1.002,
      "step": 5056
    },
    {
      "epoch": 8.70253164556962,
      "grad_norm": 0.5833732485771179,
      "learning_rate": 6.487341772151899e-06,
      "loss": 0.436,
      "step": 5500
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8830734966592427,
      "eval_loss": 0.4109877347946167,
      "eval_runtime": 71.1738,
      "eval_samples_per_second": 63.085,
      "eval_steps_per_second": 0.998,
      "step": 5688
    }
  ],
  "logging_steps": 500,
  "max_steps": 6320,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 10000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.817468380402483e+16,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}