last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0021448871253150303,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00021448871253150302,
      "grad_norm": Infinity,
      "learning_rate": 0.0,
      "loss": 11.14,
      "step": 1
    },
    {
      "epoch": 0.00021448871253150302,
      "eval_loss": 11.310040473937988,
      "eval_runtime": 238.0023,
      "eval_samples_per_second": 4.126,
      "eval_steps_per_second": 4.126,
      "step": 1
    },
    {
      "epoch": 0.00042897742506300604,
      "grad_norm": 63.52291488647461,
      "learning_rate": 2e-05,
      "loss": 11.7604,
      "step": 2
    },
    {
      "epoch": 0.0006434661375945091,
      "grad_norm": 71.64099884033203,
      "learning_rate": 4e-05,
      "loss": 13.3246,
      "step": 3
    },
    {
      "epoch": 0.0006434661375945091,
      "eval_loss": 8.893160820007324,
      "eval_runtime": 231.8354,
      "eval_samples_per_second": 4.236,
      "eval_steps_per_second": 4.236,
      "step": 3
    },
    {
      "epoch": 0.0008579548501260121,
      "grad_norm": 59.29423904418945,
      "learning_rate": 6e-05,
      "loss": 8.4669,
      "step": 4
    },
    {
      "epoch": 0.0010724435626575152,
      "grad_norm": 43.15823745727539,
      "learning_rate": 8e-05,
      "loss": 5.8545,
      "step": 5
    },
    {
      "epoch": 0.0012869322751890182,
      "grad_norm": 42.88001251220703,
      "learning_rate": 0.0001,
      "loss": 4.0161,
      "step": 6
    },
    {
      "epoch": 0.0012869322751890182,
      "eval_loss": 1.5954408645629883,
      "eval_runtime": 230.2213,
      "eval_samples_per_second": 4.265,
      "eval_steps_per_second": 4.265,
      "step": 6
    },
    {
      "epoch": 0.0015014209877205213,
      "grad_norm": 40.18137741088867,
      "learning_rate": 0.00012,
      "loss": 1.7633,
      "step": 7
    },
    {
      "epoch": 0.0017159097002520242,
      "grad_norm": 27.18015480041504,
      "learning_rate": 0.00014,
      "loss": 0.5389,
      "step": 8
    },
    {
      "epoch": 0.0019303984127835272,
      "grad_norm": 59.13418197631836,
      "learning_rate": 0.00016,
      "loss": 2.1024,
      "step": 9
    },
    {
      "epoch": 0.0019303984127835272,
      "eval_loss": 2.6004724502563477,
      "eval_runtime": 230.3217,
      "eval_samples_per_second": 4.264,
      "eval_steps_per_second": 4.264,
      "step": 9
    },
    {
      "epoch": 0.0021448871253150303,
      "grad_norm": 33.359867095947266,
      "learning_rate": 0.00018,
      "loss": 2.2976,
      "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": 7668943615426560.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}