last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00011673242593327574,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.1673242593327574e-05,
      "grad_norm": 5.800905704498291,
      "learning_rate": 2e-05,
      "loss": 3.0267,
      "step": 1
    },
    {
      "epoch": 1.1673242593327574e-05,
      "eval_loss": 4.362671852111816,
      "eval_runtime": 2660.7317,
      "eval_samples_per_second": 13.556,
      "eval_steps_per_second": 6.778,
      "step": 1
    },
    {
      "epoch": 2.3346485186655148e-05,
      "grad_norm": 6.727754592895508,
      "learning_rate": 4e-05,
      "loss": 3.9057,
      "step": 2
    },
    {
      "epoch": 3.501972777998272e-05,
      "grad_norm": 6.595926284790039,
      "learning_rate": 6e-05,
      "loss": 5.3046,
      "step": 3
    },
    {
      "epoch": 3.501972777998272e-05,
      "eval_loss": 4.294731616973877,
      "eval_runtime": 2685.7969,
      "eval_samples_per_second": 13.43,
      "eval_steps_per_second": 6.715,
      "step": 3
    },
    {
      "epoch": 4.6692970373310296e-05,
      "grad_norm": 7.166476249694824,
      "learning_rate": 8e-05,
      "loss": 3.3936,
      "step": 4
    },
    {
      "epoch": 5.836621296663787e-05,
      "grad_norm": 7.10786247253418,
      "learning_rate": 0.0001,
      "loss": 4.4623,
      "step": 5
    },
    {
      "epoch": 7.003945555996544e-05,
      "grad_norm": 7.141903400421143,
      "learning_rate": 0.00012,
      "loss": 4.1837,
      "step": 6
    },
    {
      "epoch": 7.003945555996544e-05,
      "eval_loss": 3.633279323577881,
      "eval_runtime": 2661.1457,
      "eval_samples_per_second": 13.554,
      "eval_steps_per_second": 6.777,
      "step": 6
    },
    {
      "epoch": 8.171269815329302e-05,
      "grad_norm": 6.491903781890869,
      "learning_rate": 0.00014,
      "loss": 3.2719,
      "step": 7
    },
    {
      "epoch": 9.338594074662059e-05,
      "grad_norm": 5.3933892250061035,
      "learning_rate": 0.00016,
      "loss": 2.9401,
      "step": 8
    },
    {
      "epoch": 0.00010505918333994817,
      "grad_norm": 5.092551231384277,
      "learning_rate": 0.00018,
      "loss": 1.4722,
      "step": 9
    },
    {
      "epoch": 0.00010505918333994817,
      "eval_loss": 2.5769283771514893,
      "eval_runtime": 2659.7191,
      "eval_samples_per_second": 13.562,
      "eval_steps_per_second": 6.781,
      "step": 9
    },
    {
      "epoch": 0.00011673242593327574,
      "grad_norm": 6.351680278778076,
      "learning_rate": 0.0002,
      "loss": 1.3756,
      "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": 3283829787721728.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}