last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0033472803347280333,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00033472803347280337,
      "grad_norm": 2.250663995742798,
      "learning_rate": 2e-05,
      "loss": 2.4079,
      "step": 1
    },
    {
      "epoch": 0.00033472803347280337,
      "eval_loss": 2.1328611373901367,
      "eval_runtime": 63.3813,
      "eval_samples_per_second": 19.848,
      "eval_steps_per_second": 9.924,
      "step": 1
    },
    {
      "epoch": 0.0006694560669456067,
      "grad_norm": 2.0919368267059326,
      "learning_rate": 4e-05,
      "loss": 2.1558,
      "step": 2
    },
    {
      "epoch": 0.0010041841004184101,
      "grad_norm": 1.992983102798462,
      "learning_rate": 6e-05,
      "loss": 1.9693,
      "step": 3
    },
    {
      "epoch": 0.0010041841004184101,
      "eval_loss": 2.111905336380005,
      "eval_runtime": 63.5288,
      "eval_samples_per_second": 19.802,
      "eval_steps_per_second": 9.901,
      "step": 3
    },
    {
      "epoch": 0.0013389121338912135,
      "grad_norm": 4.534268856048584,
      "learning_rate": 8e-05,
      "loss": 2.8791,
      "step": 4
    },
    {
      "epoch": 0.0016736401673640166,
      "grad_norm": 1.825005054473877,
      "learning_rate": 0.0001,
      "loss": 1.3808,
      "step": 5
    },
    {
      "epoch": 0.0020083682008368202,
      "grad_norm": 2.9952549934387207,
      "learning_rate": 0.00012,
      "loss": 1.9293,
      "step": 6
    },
    {
      "epoch": 0.0020083682008368202,
      "eval_loss": 1.9323680400848389,
      "eval_runtime": 63.5287,
      "eval_samples_per_second": 19.802,
      "eval_steps_per_second": 9.901,
      "step": 6
    },
    {
      "epoch": 0.0023430962343096236,
      "grad_norm": 2.6856002807617188,
      "learning_rate": 0.00014,
      "loss": 1.6423,
      "step": 7
    },
    {
      "epoch": 0.002677824267782427,
      "grad_norm": 2.435944080352783,
      "learning_rate": 0.00016,
      "loss": 1.9694,
      "step": 8
    },
    {
      "epoch": 0.0030125523012552303,
      "grad_norm": 2.048276901245117,
      "learning_rate": 0.00018,
      "loss": 1.5035,
      "step": 9
    },
    {
      "epoch": 0.0030125523012552303,
      "eval_loss": 1.608821153640747,
      "eval_runtime": 63.5361,
      "eval_samples_per_second": 19.8,
      "eval_steps_per_second": 9.9,
      "step": 9
    },
    {
      "epoch": 0.0033472803347280333,
      "grad_norm": 1.8150256872177124,
      "learning_rate": 0.0002,
      "loss": 1.8002,
      "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": 2962191695216640.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}