last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.005961251862891207,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005961251862891207,
      "grad_norm": 14.602910041809082,
      "learning_rate": 2e-05,
      "loss": 10.8252,
      "step": 1
    },
    {
      "epoch": 0.0005961251862891207,
      "eval_loss": 11.547547340393066,
      "eval_runtime": 13.0684,
      "eval_samples_per_second": 54.1,
      "eval_steps_per_second": 27.088,
      "step": 1
    },
    {
      "epoch": 0.0011922503725782414,
      "grad_norm": 15.501795768737793,
      "learning_rate": 4e-05,
      "loss": 10.6857,
      "step": 2
    },
    {
      "epoch": 0.0017883755588673621,
      "grad_norm": 14.309174537658691,
      "learning_rate": 6e-05,
      "loss": 12.2687,
      "step": 3
    },
    {
      "epoch": 0.0017883755588673621,
      "eval_loss": 11.453269004821777,
      "eval_runtime": 12.9917,
      "eval_samples_per_second": 54.419,
      "eval_steps_per_second": 27.248,
      "step": 3
    },
    {
      "epoch": 0.002384500745156483,
      "grad_norm": 14.269388198852539,
      "learning_rate": 8e-05,
      "loss": 10.821,
      "step": 4
    },
    {
      "epoch": 0.0029806259314456036,
      "grad_norm": 17.612112045288086,
      "learning_rate": 0.0001,
      "loss": 12.8865,
      "step": 5
    },
    {
      "epoch": 0.0035767511177347243,
      "grad_norm": 18.52869415283203,
      "learning_rate": 0.00012,
      "loss": 10.7903,
      "step": 6
    },
    {
      "epoch": 0.0035767511177347243,
      "eval_loss": 9.825249671936035,
      "eval_runtime": 13.2456,
      "eval_samples_per_second": 53.376,
      "eval_steps_per_second": 26.726,
      "step": 6
    },
    {
      "epoch": 0.004172876304023845,
      "grad_norm": 23.56140899658203,
      "learning_rate": 0.00014,
      "loss": 9.075,
      "step": 7
    },
    {
      "epoch": 0.004769001490312966,
      "grad_norm": 23.681167602539062,
      "learning_rate": 0.00016,
      "loss": 8.3211,
      "step": 8
    },
    {
      "epoch": 0.005365126676602086,
      "grad_norm": 28.13359832763672,
      "learning_rate": 0.00018,
      "loss": 7.8672,
      "step": 9
    },
    {
      "epoch": 0.005365126676602086,
      "eval_loss": 4.785869598388672,
      "eval_runtime": 13.1762,
      "eval_samples_per_second": 53.658,
      "eval_steps_per_second": 26.867,
      "step": 9
    },
    {
      "epoch": 0.005961251862891207,
      "grad_norm": 28.6697998046875,
      "learning_rate": 0.0002,
      "loss": 4.9506,
      "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": 501714738216960.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}