last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0007514418290094118,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 7.514418290094118e-05,
      "grad_norm": 1.1709792613983154,
      "learning_rate": 2e-05,
      "loss": 3.1717,
      "step": 1
    },
    {
      "epoch": 7.514418290094118e-05,
      "eval_loss": 3.02657413482666,
      "eval_runtime": 3626.5488,
      "eval_samples_per_second": 1.545,
      "eval_steps_per_second": 0.773,
      "step": 1
    },
    {
      "epoch": 0.00015028836580188235,
      "grad_norm": 1.1818255186080933,
      "learning_rate": 4e-05,
      "loss": 3.1186,
      "step": 2
    },
    {
      "epoch": 0.00022543254870282354,
      "grad_norm": 1.1788246631622314,
      "learning_rate": 6e-05,
      "loss": 3.0077,
      "step": 3
    },
    {
      "epoch": 0.00022543254870282354,
      "eval_loss": 3.011431932449341,
      "eval_runtime": 3716.1993,
      "eval_samples_per_second": 1.508,
      "eval_steps_per_second": 0.754,
      "step": 3
    },
    {
      "epoch": 0.0003005767316037647,
      "grad_norm": 1.2218079566955566,
      "learning_rate": 8e-05,
      "loss": 3.0134,
      "step": 4
    },
    {
      "epoch": 0.0003757209145047059,
      "grad_norm": 1.3877516984939575,
      "learning_rate": 0.0001,
      "loss": 2.8411,
      "step": 5
    },
    {
      "epoch": 0.0004508650974056471,
      "grad_norm": 1.505188226699829,
      "learning_rate": 0.00012,
      "loss": 2.6962,
      "step": 6
    },
    {
      "epoch": 0.0004508650974056471,
      "eval_loss": 2.8419442176818848,
      "eval_runtime": 3637.7307,
      "eval_samples_per_second": 1.541,
      "eval_steps_per_second": 0.77,
      "step": 6
    },
    {
      "epoch": 0.0005260092803065883,
      "grad_norm": 0.9824367761611938,
      "learning_rate": 0.00014,
      "loss": 2.9446,
      "step": 7
    },
    {
      "epoch": 0.0006011534632075294,
      "grad_norm": 1.1818082332611084,
      "learning_rate": 0.00016,
      "loss": 3.0576,
      "step": 8
    },
    {
      "epoch": 0.0006762976461084706,
      "grad_norm": 1.307089924812317,
      "learning_rate": 0.00018,
      "loss": 2.4956,
      "step": 9
    },
    {
      "epoch": 0.0006762976461084706,
      "eval_loss": 2.709524631500244,
      "eval_runtime": 3641.0785,
      "eval_samples_per_second": 1.539,
      "eval_steps_per_second": 0.77,
      "step": 9
    },
    {
      "epoch": 0.0007514418290094118,
      "grad_norm": 1.2753934860229492,
      "learning_rate": 0.0002,
      "loss": 2.8668,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.483774567120896e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}