checkpoint-1125/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1125,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 0.2409825623035431,
      "learning_rate": 0.0001,
      "loss": 1.7886,
      "step": 100
    },
    {
      "epoch": 0.9973333333333333,
      "eval_accuracy": 0.6061076233183856,
      "eval_loss": 1.6901254653930664,
      "eval_runtime": 5.5484,
      "eval_samples_per_second": 90.116,
      "eval_steps_per_second": 11.355,
      "step": 187
    },
    {
      "epoch": 1.0666666666666667,
      "grad_norm": 0.2729445993900299,
      "learning_rate": 0.0001,
      "loss": 1.7283,
      "step": 200
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.37678101658821106,
      "learning_rate": 0.0001,
      "loss": 1.6544,
      "step": 300
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.607677130044843,
      "eval_loss": 1.6766204833984375,
      "eval_runtime": 6.4294,
      "eval_samples_per_second": 77.768,
      "eval_steps_per_second": 9.799,
      "step": 375
    },
    {
      "epoch": 2.1333333333333333,
      "grad_norm": 0.4604090750217438,
      "learning_rate": 0.0001,
      "loss": 1.6338,
      "step": 400
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.5606330633163452,
      "learning_rate": 0.0001,
      "loss": 1.5273,
      "step": 500
    },
    {
      "epoch": 2.997333333333333,
      "eval_accuracy": 0.6079910313901346,
      "eval_loss": 1.6929064989089966,
      "eval_runtime": 5.431,
      "eval_samples_per_second": 92.063,
      "eval_steps_per_second": 11.6,
      "step": 562
    },
    {
      "epoch": 3.2,
      "grad_norm": 0.6363856196403503,
      "learning_rate": 0.0001,
      "loss": 1.4796,
      "step": 600
    },
    {
      "epoch": 3.7333333333333334,
      "grad_norm": 0.5985731482505798,
      "learning_rate": 0.0001,
      "loss": 1.3871,
      "step": 700
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.606896860986547,
      "eval_loss": 1.7257370948791504,
      "eval_runtime": 5.8318,
      "eval_samples_per_second": 85.737,
      "eval_steps_per_second": 10.803,
      "step": 750
    },
    {
      "epoch": 4.266666666666667,
      "grad_norm": 0.7769015431404114,
      "learning_rate": 0.0001,
      "loss": 1.3013,
      "step": 800
    },
    {
      "epoch": 4.8,
      "grad_norm": 0.708043098449707,
      "learning_rate": 0.0001,
      "loss": 1.23,
      "step": 900
    },
    {
      "epoch": 4.997333333333334,
      "eval_accuracy": 0.6061076233183856,
      "eval_loss": 1.7812739610671997,
      "eval_runtime": 5.9335,
      "eval_samples_per_second": 84.267,
      "eval_steps_per_second": 10.618,
      "step": 937
    },
    {
      "epoch": 5.333333333333333,
      "grad_norm": 0.8580867648124695,
      "learning_rate": 0.0001,
      "loss": 1.1011,
      "step": 1000
    },
    {
      "epoch": 5.866666666666667,
      "grad_norm": 0.9232878684997559,
      "learning_rate": 0.0001,
      "loss": 1.0749,
      "step": 1100
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6018026905829597,
      "eval_loss": 1.877648949623108,
      "eval_runtime": 6.474,
      "eval_samples_per_second": 77.231,
      "eval_steps_per_second": 9.731,
      "step": 1125
    }
  ],
  "logging_steps": 100,
  "max_steps": 9350,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.1331927290989773e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}