last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0016923337282112032,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00016923337282112033,
      "grad_norm": 2.015991449356079,
      "learning_rate": 2e-05,
      "loss": 9.1743,
      "step": 1
    },
    {
      "epoch": 0.00016923337282112033,
      "eval_loss": 2.541017532348633,
      "eval_runtime": 260.1039,
      "eval_samples_per_second": 9.569,
      "eval_steps_per_second": 4.787,
      "step": 1
    },
    {
      "epoch": 0.00033846674564224065,
      "grad_norm": 2.6391468048095703,
      "learning_rate": 4e-05,
      "loss": 9.9181,
      "step": 2
    },
    {
      "epoch": 0.000507700118463361,
      "grad_norm": 3.450047254562378,
      "learning_rate": 6e-05,
      "loss": 10.6527,
      "step": 3
    },
    {
      "epoch": 0.000507700118463361,
      "eval_loss": 2.538522720336914,
      "eval_runtime": 259.9061,
      "eval_samples_per_second": 9.577,
      "eval_steps_per_second": 4.79,
      "step": 3
    },
    {
      "epoch": 0.0006769334912844813,
      "grad_norm": 3.2111623287200928,
      "learning_rate": 8e-05,
      "loss": 10.3007,
      "step": 4
    },
    {
      "epoch": 0.0008461668641056016,
      "grad_norm": 3.6762893199920654,
      "learning_rate": 0.0001,
      "loss": 8.6931,
      "step": 5
    },
    {
      "epoch": 0.001015400236926722,
      "grad_norm": 4.102634906768799,
      "learning_rate": 0.00012,
      "loss": 9.1596,
      "step": 6
    },
    {
      "epoch": 0.001015400236926722,
      "eval_loss": 2.4958791732788086,
      "eval_runtime": 260.0942,
      "eval_samples_per_second": 9.57,
      "eval_steps_per_second": 4.787,
      "step": 6
    },
    {
      "epoch": 0.0011846336097478423,
      "grad_norm": 2.7566802501678467,
      "learning_rate": 0.00014,
      "loss": 7.4648,
      "step": 7
    },
    {
      "epoch": 0.0013538669825689626,
      "grad_norm": 3.6192877292633057,
      "learning_rate": 0.00016,
      "loss": 8.6464,
      "step": 8
    },
    {
      "epoch": 0.001523100355390083,
      "grad_norm": 7.141613483428955,
      "learning_rate": 0.00018,
      "loss": 10.1607,
      "step": 9
    },
    {
      "epoch": 0.001523100355390083,
      "eval_loss": 2.2586708068847656,
      "eval_runtime": 260.1385,
      "eval_samples_per_second": 9.568,
      "eval_steps_per_second": 4.786,
      "step": 9
    },
    {
      "epoch": 0.0016923337282112032,
      "grad_norm": 7.445590496063232,
      "learning_rate": 0.0002,
      "loss": 9.6125,
      "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": 3166224305356800.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}