last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.003437016669530847,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00034370166695308474,
      "grad_norm": 0.5219367146492004,
      "learning_rate": 2e-05,
      "loss": 1.7053,
      "step": 1
    },
    {
      "epoch": 0.00034370166695308474,
      "eval_loss": 1.7383819818496704,
      "eval_runtime": 75.8139,
      "eval_samples_per_second": 16.171,
      "eval_steps_per_second": 8.086,
      "step": 1
    },
    {
      "epoch": 0.0006874033339061695,
      "grad_norm": 0.5793853402137756,
      "learning_rate": 4e-05,
      "loss": 1.8144,
      "step": 2
    },
    {
      "epoch": 0.0010311050008592541,
      "grad_norm": 0.4772932529449463,
      "learning_rate": 6e-05,
      "loss": 1.7517,
      "step": 3
    },
    {
      "epoch": 0.0010311050008592541,
      "eval_loss": 1.732722520828247,
      "eval_runtime": 75.3966,
      "eval_samples_per_second": 16.261,
      "eval_steps_per_second": 8.13,
      "step": 3
    },
    {
      "epoch": 0.001374806667812339,
      "grad_norm": 0.5124571323394775,
      "learning_rate": 8e-05,
      "loss": 1.6873,
      "step": 4
    },
    {
      "epoch": 0.0017185083347654236,
      "grad_norm": 0.5481643676757812,
      "learning_rate": 0.0001,
      "loss": 1.7401,
      "step": 5
    },
    {
      "epoch": 0.0020622100017185082,
      "grad_norm": 0.6607972383499146,
      "learning_rate": 0.00012,
      "loss": 1.6403,
      "step": 6
    },
    {
      "epoch": 0.0020622100017185082,
      "eval_loss": 1.6614868640899658,
      "eval_runtime": 75.7943,
      "eval_samples_per_second": 16.175,
      "eval_steps_per_second": 8.088,
      "step": 6
    },
    {
      "epoch": 0.002405911668671593,
      "grad_norm": 0.7821069359779358,
      "learning_rate": 0.00014,
      "loss": 1.5828,
      "step": 7
    },
    {
      "epoch": 0.002749613335624678,
      "grad_norm": 0.7105597853660583,
      "learning_rate": 0.00016,
      "loss": 1.5453,
      "step": 8
    },
    {
      "epoch": 0.0030933150025777623,
      "grad_norm": 0.6306270360946655,
      "learning_rate": 0.00018,
      "loss": 1.4635,
      "step": 9
    },
    {
      "epoch": 0.0030933150025777623,
      "eval_loss": 1.4273592233657837,
      "eval_runtime": 75.5366,
      "eval_samples_per_second": 16.231,
      "eval_steps_per_second": 8.115,
      "step": 9
    },
    {
      "epoch": 0.003437016669530847,
      "grad_norm": 0.581282913684845,
      "learning_rate": 0.0002,
      "loss": 1.3731,
      "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": 1628733135912960.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}