last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.010598834128245893,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0010598834128245894,
      "grad_norm": 0.6350849270820618,
      "learning_rate": 2e-05,
      "loss": 5.9436,
      "step": 1
    },
    {
      "epoch": 0.0010598834128245894,
      "eval_loss": 1.5727260112762451,
      "eval_runtime": 27.5042,
      "eval_samples_per_second": 14.471,
      "eval_steps_per_second": 7.235,
      "step": 1
    },
    {
      "epoch": 0.0021197668256491787,
      "grad_norm": 0.601396918296814,
      "learning_rate": 4e-05,
      "loss": 7.3717,
      "step": 2
    },
    {
      "epoch": 0.003179650238473768,
      "grad_norm": 0.6531712412834167,
      "learning_rate": 6e-05,
      "loss": 6.0098,
      "step": 3
    },
    {
      "epoch": 0.003179650238473768,
      "eval_loss": 1.5714995861053467,
      "eval_runtime": 27.5049,
      "eval_samples_per_second": 14.47,
      "eval_steps_per_second": 7.235,
      "step": 3
    },
    {
      "epoch": 0.0042395336512983575,
      "grad_norm": 0.7442638278007507,
      "learning_rate": 8e-05,
      "loss": 6.4499,
      "step": 4
    },
    {
      "epoch": 0.005299417064122946,
      "grad_norm": 0.7729753851890564,
      "learning_rate": 0.0001,
      "loss": 6.1611,
      "step": 5
    },
    {
      "epoch": 0.006359300476947536,
      "grad_norm": 0.7975832223892212,
      "learning_rate": 0.00012,
      "loss": 6.1086,
      "step": 6
    },
    {
      "epoch": 0.006359300476947536,
      "eval_loss": 1.549599051475525,
      "eval_runtime": 27.5431,
      "eval_samples_per_second": 14.45,
      "eval_steps_per_second": 7.225,
      "step": 6
    },
    {
      "epoch": 0.007419183889772125,
      "grad_norm": 0.9927419424057007,
      "learning_rate": 0.00014,
      "loss": 6.2598,
      "step": 7
    },
    {
      "epoch": 0.008479067302596715,
      "grad_norm": 1.246896743774414,
      "learning_rate": 0.00016,
      "loss": 6.0118,
      "step": 8
    },
    {
      "epoch": 0.009538950715421303,
      "grad_norm": 1.5003572702407837,
      "learning_rate": 0.00018,
      "loss": 5.9791,
      "step": 9
    },
    {
      "epoch": 0.009538950715421303,
      "eval_loss": 1.4221528768539429,
      "eval_runtime": 27.5518,
      "eval_samples_per_second": 14.446,
      "eval_steps_per_second": 7.223,
      "step": 9
    },
    {
      "epoch": 0.010598834128245893,
      "grad_norm": 1.3482239246368408,
      "learning_rate": 0.0002,
      "loss": 5.7713,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3257498484080640.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}