trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.981366459627329,
  "eval_steps": 20,
  "global_step": 160,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.012422360248447204,
      "grad_norm": 0.1855994164943695,
      "learning_rate": 0.0005,
      "loss": 0.7756,
      "step": 1
    },
    {
      "epoch": 0.2484472049689441,
      "grad_norm": 0.300306499004364,
      "learning_rate": 0.0005,
      "loss": 0.6307,
      "step": 20
    },
    {
      "epoch": 0.2484472049689441,
      "eval_loss": 0.5943351984024048,
      "eval_runtime": 244.8595,
      "eval_samples_per_second": 0.314,
      "eval_steps_per_second": 0.314,
      "step": 20
    },
    {
      "epoch": 0.4968944099378882,
      "grad_norm": 0.21684305369853973,
      "learning_rate": 0.0005,
      "loss": 0.5783,
      "step": 40
    },
    {
      "epoch": 0.4968944099378882,
      "eval_loss": 0.5758653879165649,
      "eval_runtime": 245.0528,
      "eval_samples_per_second": 0.314,
      "eval_steps_per_second": 0.314,
      "step": 40
    },
    {
      "epoch": 0.7453416149068323,
      "grad_norm": 0.3364117741584778,
      "learning_rate": 0.0005,
      "loss": 0.5785,
      "step": 60
    },
    {
      "epoch": 0.7453416149068323,
      "eval_loss": 0.5718451738357544,
      "eval_runtime": 244.9796,
      "eval_samples_per_second": 0.314,
      "eval_steps_per_second": 0.314,
      "step": 60
    },
    {
      "epoch": 0.9937888198757764,
      "grad_norm": 0.31104227900505066,
      "learning_rate": 0.0005,
      "loss": 0.5789,
      "step": 80
    },
    {
      "epoch": 0.9937888198757764,
      "eval_loss": 0.5634680986404419,
      "eval_runtime": 244.8128,
      "eval_samples_per_second": 0.315,
      "eval_steps_per_second": 0.315,
      "step": 80
    },
    {
      "epoch": 1.236024844720497,
      "grad_norm": 0.29026365280151367,
      "learning_rate": 0.0005,
      "loss": 0.4537,
      "step": 100
    },
    {
      "epoch": 1.236024844720497,
      "eval_loss": 0.5781797170639038,
      "eval_runtime": 245.0239,
      "eval_samples_per_second": 0.314,
      "eval_steps_per_second": 0.314,
      "step": 100
    },
    {
      "epoch": 1.484472049689441,
      "grad_norm": 0.3176882266998291,
      "learning_rate": 0.0005,
      "loss": 0.4573,
      "step": 120
    },
    {
      "epoch": 1.484472049689441,
      "eval_loss": 0.5743211507797241,
      "eval_runtime": 244.9113,
      "eval_samples_per_second": 0.314,
      "eval_steps_per_second": 0.314,
      "step": 120
    },
    {
      "epoch": 1.7329192546583851,
      "grad_norm": 0.35556861758232117,
      "learning_rate": 0.0005,
      "loss": 0.4619,
      "step": 140
    },
    {
      "epoch": 1.7329192546583851,
      "eval_loss": 0.5818625092506409,
      "eval_runtime": 245.1195,
      "eval_samples_per_second": 0.314,
      "eval_steps_per_second": 0.314,
      "step": 140
    },
    {
      "epoch": 1.981366459627329,
      "grad_norm": 0.37015023827552795,
      "learning_rate": 0.0005,
      "loss": 0.4689,
      "step": 160
    },
    {
      "epoch": 1.981366459627329,
      "eval_loss": 0.5779598355293274,
      "eval_runtime": 245.0391,
      "eval_samples_per_second": 0.314,
      "eval_steps_per_second": 0.314,
      "step": 160
    }
  ],
  "logging_steps": 20,
  "max_steps": 240,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 20,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 7.328341998899405e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}