last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0010707784559374665,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00010707784559374665,
      "grad_norm": 0.5286692380905151,
      "learning_rate": 2e-05,
      "loss": 1.0893,
      "step": 1
    },
    {
      "epoch": 0.00010707784559374665,
      "eval_loss": 1.2124216556549072,
      "eval_runtime": 186.8738,
      "eval_samples_per_second": 10.526,
      "eval_steps_per_second": 10.526,
      "step": 1
    },
    {
      "epoch": 0.0002141556911874933,
      "grad_norm": 0.6973130106925964,
      "learning_rate": 4e-05,
      "loss": 1.1445,
      "step": 2
    },
    {
      "epoch": 0.00032123353678124,
      "grad_norm": 0.8045721054077148,
      "learning_rate": 6e-05,
      "loss": 1.2625,
      "step": 3
    },
    {
      "epoch": 0.00032123353678124,
      "eval_loss": 1.1848317384719849,
      "eval_runtime": 183.2207,
      "eval_samples_per_second": 10.736,
      "eval_steps_per_second": 10.736,
      "step": 3
    },
    {
      "epoch": 0.0004283113823749866,
      "grad_norm": 0.5536651015281677,
      "learning_rate": 8e-05,
      "loss": 1.1659,
      "step": 4
    },
    {
      "epoch": 0.0005353892279687332,
      "grad_norm": 0.4006960093975067,
      "learning_rate": 0.0001,
      "loss": 1.0878,
      "step": 5
    },
    {
      "epoch": 0.00064246707356248,
      "grad_norm": 0.34919866919517517,
      "learning_rate": 0.00012,
      "loss": 1.0365,
      "step": 6
    },
    {
      "epoch": 0.00064246707356248,
      "eval_loss": 1.0633049011230469,
      "eval_runtime": 184.7088,
      "eval_samples_per_second": 10.649,
      "eval_steps_per_second": 10.649,
      "step": 6
    },
    {
      "epoch": 0.0007495449191562266,
      "grad_norm": 0.36104100942611694,
      "learning_rate": 0.00014,
      "loss": 0.9956,
      "step": 7
    },
    {
      "epoch": 0.0008566227647499732,
      "grad_norm": 0.4580983519554138,
      "learning_rate": 0.00016,
      "loss": 1.0359,
      "step": 8
    },
    {
      "epoch": 0.0009637006103437199,
      "grad_norm": 0.5145609974861145,
      "learning_rate": 0.00018,
      "loss": 0.8883,
      "step": 9
    },
    {
      "epoch": 0.0009637006103437199,
      "eval_loss": 0.8663086295127869,
      "eval_runtime": 184.9839,
      "eval_samples_per_second": 10.633,
      "eval_steps_per_second": 10.633,
      "step": 9
    },
    {
      "epoch": 0.0010707784559374665,
      "grad_norm": 0.6000295877456665,
      "learning_rate": 0.0002,
      "loss": 0.7515,
      "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": 2768846331052032.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}