last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004253056884635832,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0004253056884635832,
      "grad_norm": 0.6765125393867493,
      "learning_rate": 2e-05,
      "loss": 1.7296,
      "step": 1
    },
    {
      "epoch": 0.0004253056884635832,
      "eval_loss": 1.7417739629745483,
      "eval_runtime": 13.6849,
      "eval_samples_per_second": 72.343,
      "eval_steps_per_second": 36.171,
      "step": 1
    },
    {
      "epoch": 0.0008506113769271664,
      "grad_norm": 0.6839314103126526,
      "learning_rate": 4e-05,
      "loss": 1.8076,
      "step": 2
    },
    {
      "epoch": 0.0012759170653907496,
      "grad_norm": 0.6799637079238892,
      "learning_rate": 6e-05,
      "loss": 1.7601,
      "step": 3
    },
    {
      "epoch": 0.0012759170653907496,
      "eval_loss": 1.738396167755127,
      "eval_runtime": 13.1982,
      "eval_samples_per_second": 75.01,
      "eval_steps_per_second": 37.505,
      "step": 3
    },
    {
      "epoch": 0.0017012227538543328,
      "grad_norm": 0.7580278515815735,
      "learning_rate": 8e-05,
      "loss": 1.7209,
      "step": 4
    },
    {
      "epoch": 0.002126528442317916,
      "grad_norm": 0.8269679546356201,
      "learning_rate": 0.0001,
      "loss": 1.7431,
      "step": 5
    },
    {
      "epoch": 0.002551834130781499,
      "grad_norm": 0.6974154710769653,
      "learning_rate": 0.00012,
      "loss": 1.7149,
      "step": 6
    },
    {
      "epoch": 0.002551834130781499,
      "eval_loss": 1.6992298364639282,
      "eval_runtime": 13.3487,
      "eval_samples_per_second": 74.165,
      "eval_steps_per_second": 37.082,
      "step": 6
    },
    {
      "epoch": 0.0029771398192450825,
      "grad_norm": 0.8422204852104187,
      "learning_rate": 0.00014,
      "loss": 1.8683,
      "step": 7
    },
    {
      "epoch": 0.0034024455077086655,
      "grad_norm": 0.7477498650550842,
      "learning_rate": 0.00016,
      "loss": 1.5238,
      "step": 8
    },
    {
      "epoch": 0.003827751196172249,
      "grad_norm": 0.7416492104530334,
      "learning_rate": 0.00018,
      "loss": 1.6067,
      "step": 9
    },
    {
      "epoch": 0.003827751196172249,
      "eval_loss": 1.580859661102295,
      "eval_runtime": 13.3094,
      "eval_samples_per_second": 74.384,
      "eval_steps_per_second": 37.192,
      "step": 9
    },
    {
      "epoch": 0.004253056884635832,
      "grad_norm": 0.8465039134025574,
      "learning_rate": 0.0002,
      "loss": 1.6505,
      "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": 255791978250240.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}