last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00015873645779594428,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.587364577959443e-05,
      "grad_norm": 1.7801799774169922,
      "learning_rate": 2e-05,
      "loss": 1.7349,
      "step": 1
    },
    {
      "epoch": 1.587364577959443e-05,
      "eval_loss": 2.8036000728607178,
      "eval_runtime": 587.2782,
      "eval_samples_per_second": 45.168,
      "eval_steps_per_second": 22.584,
      "step": 1
    },
    {
      "epoch": 3.174729155918886e-05,
      "grad_norm": 2.7413840293884277,
      "learning_rate": 4e-05,
      "loss": 2.3991,
      "step": 2
    },
    {
      "epoch": 4.762093733878329e-05,
      "grad_norm": 2.748443841934204,
      "learning_rate": 6e-05,
      "loss": 3.015,
      "step": 3
    },
    {
      "epoch": 4.762093733878329e-05,
      "eval_loss": 2.7955071926116943,
      "eval_runtime": 506.2626,
      "eval_samples_per_second": 52.396,
      "eval_steps_per_second": 26.198,
      "step": 3
    },
    {
      "epoch": 6.349458311837772e-05,
      "grad_norm": 1.9895458221435547,
      "learning_rate": 8e-05,
      "loss": 2.449,
      "step": 4
    },
    {
      "epoch": 7.936822889797214e-05,
      "grad_norm": 1.320906162261963,
      "learning_rate": 0.0001,
      "loss": 2.5093,
      "step": 5
    },
    {
      "epoch": 9.524187467756657e-05,
      "grad_norm": 2.114384889602661,
      "learning_rate": 0.00012,
      "loss": 2.451,
      "step": 6
    },
    {
      "epoch": 9.524187467756657e-05,
      "eval_loss": 2.7090823650360107,
      "eval_runtime": 520.5763,
      "eval_samples_per_second": 50.955,
      "eval_steps_per_second": 25.478,
      "step": 6
    },
    {
      "epoch": 0.000111115520457161,
      "grad_norm": 2.729943037033081,
      "learning_rate": 0.00014,
      "loss": 3.3698,
      "step": 7
    },
    {
      "epoch": 0.00012698916623675544,
      "grad_norm": 1.4825519323349,
      "learning_rate": 0.00016,
      "loss": 2.4287,
      "step": 8
    },
    {
      "epoch": 0.00014286281201634986,
      "grad_norm": 1.5584574937820435,
      "learning_rate": 0.00018,
      "loss": 2.221,
      "step": 9
    },
    {
      "epoch": 0.00014286281201634986,
      "eval_loss": 2.539081573486328,
      "eval_runtime": 431.5948,
      "eval_samples_per_second": 61.46,
      "eval_steps_per_second": 30.73,
      "step": 9
    },
    {
      "epoch": 0.00015873645779594428,
      "grad_norm": 1.7351861000061035,
      "learning_rate": 0.0002,
      "loss": 2.3476,
      "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": 89038163804160.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}