last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00041611185086551264,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.161118508655127e-05,
      "grad_norm": 5.774013996124268,
      "learning_rate": 2e-05,
      "loss": 3.2351,
      "step": 1
    },
    {
      "epoch": 4.161118508655127e-05,
      "eval_loss": 0.8437787890434265,
      "eval_runtime": 579.8894,
      "eval_samples_per_second": 17.45,
      "eval_steps_per_second": 8.726,
      "step": 1
    },
    {
      "epoch": 8.322237017310254e-05,
      "grad_norm": 5.5636444091796875,
      "learning_rate": 4e-05,
      "loss": 3.1196,
      "step": 2
    },
    {
      "epoch": 0.0001248335552596538,
      "grad_norm": 6.209656715393066,
      "learning_rate": 6e-05,
      "loss": 3.3023,
      "step": 3
    },
    {
      "epoch": 0.0001248335552596538,
      "eval_loss": 0.835146427154541,
      "eval_runtime": 578.4735,
      "eval_samples_per_second": 17.493,
      "eval_steps_per_second": 8.747,
      "step": 3
    },
    {
      "epoch": 0.00016644474034620507,
      "grad_norm": 5.123037815093994,
      "learning_rate": 8e-05,
      "loss": 3.1748,
      "step": 4
    },
    {
      "epoch": 0.00020805592543275632,
      "grad_norm": 6.092409133911133,
      "learning_rate": 0.0001,
      "loss": 3.4404,
      "step": 5
    },
    {
      "epoch": 0.0002496671105193076,
      "grad_norm": 6.741403102874756,
      "learning_rate": 0.00012,
      "loss": 3.6998,
      "step": 6
    },
    {
      "epoch": 0.0002496671105193076,
      "eval_loss": 0.7651312351226807,
      "eval_runtime": 578.7606,
      "eval_samples_per_second": 17.484,
      "eval_steps_per_second": 8.743,
      "step": 6
    },
    {
      "epoch": 0.00029127829560585887,
      "grad_norm": 5.375231742858887,
      "learning_rate": 0.00014,
      "loss": 3.0135,
      "step": 7
    },
    {
      "epoch": 0.00033288948069241014,
      "grad_norm": 5.039647579193115,
      "learning_rate": 0.00016,
      "loss": 3.0762,
      "step": 8
    },
    {
      "epoch": 0.00037450066577896136,
      "grad_norm": 4.423862934112549,
      "learning_rate": 0.00018,
      "loss": 2.4585,
      "step": 9
    },
    {
      "epoch": 0.00037450066577896136,
      "eval_loss": 0.6939747333526611,
      "eval_runtime": 578.6516,
      "eval_samples_per_second": 17.487,
      "eval_steps_per_second": 8.744,
      "step": 9
    },
    {
      "epoch": 0.00041611185086551264,
      "grad_norm": 4.605961322784424,
      "learning_rate": 0.0002,
      "loss": 3.151,
      "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": 3505341585162240.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}