last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.07874015748031496,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.007874015748031496,
      "grad_norm": 3.6459546089172363,
      "learning_rate": 2e-05,
      "loss": 8.7273,
      "step": 1
    },
    {
      "epoch": 0.007874015748031496,
      "eval_loss": 8.648087501525879,
      "eval_runtime": 5.7219,
      "eval_samples_per_second": 9.437,
      "eval_steps_per_second": 4.719,
      "step": 1
    },
    {
      "epoch": 0.015748031496062992,
      "grad_norm": 3.5772509574890137,
      "learning_rate": 4e-05,
      "loss": 8.11,
      "step": 2
    },
    {
      "epoch": 0.023622047244094488,
      "grad_norm": 3.6938021183013916,
      "learning_rate": 6e-05,
      "loss": 8.8905,
      "step": 3
    },
    {
      "epoch": 0.023622047244094488,
      "eval_loss": 8.622480392456055,
      "eval_runtime": 5.3807,
      "eval_samples_per_second": 10.036,
      "eval_steps_per_second": 5.018,
      "step": 3
    },
    {
      "epoch": 0.031496062992125984,
      "grad_norm": 4.027983665466309,
      "learning_rate": 8e-05,
      "loss": 8.5513,
      "step": 4
    },
    {
      "epoch": 0.03937007874015748,
      "grad_norm": 4.555736541748047,
      "learning_rate": 0.0001,
      "loss": 7.9343,
      "step": 5
    },
    {
      "epoch": 0.047244094488188976,
      "grad_norm": 5.244207859039307,
      "learning_rate": 0.00012,
      "loss": 8.4064,
      "step": 6
    },
    {
      "epoch": 0.047244094488188976,
      "eval_loss": 7.6861042976379395,
      "eval_runtime": 5.4901,
      "eval_samples_per_second": 9.836,
      "eval_steps_per_second": 4.918,
      "step": 6
    },
    {
      "epoch": 0.05511811023622047,
      "grad_norm": 5.364976406097412,
      "learning_rate": 0.00014,
      "loss": 7.5692,
      "step": 7
    },
    {
      "epoch": 0.06299212598425197,
      "grad_norm": 6.0303497314453125,
      "learning_rate": 0.00016,
      "loss": 7.0766,
      "step": 8
    },
    {
      "epoch": 0.07086614173228346,
      "grad_norm": 6.065710067749023,
      "learning_rate": 0.00018,
      "loss": 5.7276,
      "step": 9
    },
    {
      "epoch": 0.07086614173228346,
      "eval_loss": 5.128417015075684,
      "eval_runtime": 5.5492,
      "eval_samples_per_second": 9.731,
      "eval_steps_per_second": 4.866,
      "step": 9
    },
    {
      "epoch": 0.07874015748031496,
      "grad_norm": 6.839983940124512,
      "learning_rate": 0.0002,
      "loss": 5.1767,
      "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": 1742636046090240.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}