last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0014972301242701003,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00014972301242701004,
      "grad_norm": 5.4886040687561035,
      "learning_rate": 2e-05,
      "loss": 13.9691,
      "step": 1
    },
    {
      "epoch": 0.00014972301242701004,
      "eval_loss": 3.1406643390655518,
      "eval_runtime": 94.6682,
      "eval_samples_per_second": 29.714,
      "eval_steps_per_second": 14.862,
      "step": 1
    },
    {
      "epoch": 0.0002994460248540201,
      "grad_norm": 3.7217674255371094,
      "learning_rate": 4e-05,
      "loss": 12.1924,
      "step": 2
    },
    {
      "epoch": 0.0004491690372810301,
      "grad_norm": 3.217331886291504,
      "learning_rate": 6e-05,
      "loss": 10.7015,
      "step": 3
    },
    {
      "epoch": 0.0004491690372810301,
      "eval_loss": 3.133810520172119,
      "eval_runtime": 94.5561,
      "eval_samples_per_second": 29.75,
      "eval_steps_per_second": 14.88,
      "step": 3
    },
    {
      "epoch": 0.0005988920497080402,
      "grad_norm": 4.119112014770508,
      "learning_rate": 8e-05,
      "loss": 13.303,
      "step": 4
    },
    {
      "epoch": 0.0007486150621350501,
      "grad_norm": 3.7569944858551025,
      "learning_rate": 0.0001,
      "loss": 12.3719,
      "step": 5
    },
    {
      "epoch": 0.0008983380745620602,
      "grad_norm": 5.156914710998535,
      "learning_rate": 0.00012,
      "loss": 13.0945,
      "step": 6
    },
    {
      "epoch": 0.0008983380745620602,
      "eval_loss": 3.020592451095581,
      "eval_runtime": 94.5439,
      "eval_samples_per_second": 29.753,
      "eval_steps_per_second": 14.882,
      "step": 6
    },
    {
      "epoch": 0.0010480610869890701,
      "grad_norm": 7.51618766784668,
      "learning_rate": 0.00014,
      "loss": 12.9868,
      "step": 7
    },
    {
      "epoch": 0.0011977840994160803,
      "grad_norm": 7.297400951385498,
      "learning_rate": 0.00016,
      "loss": 11.1761,
      "step": 8
    },
    {
      "epoch": 0.0013475071118430903,
      "grad_norm": 8.24133586883545,
      "learning_rate": 0.00018,
      "loss": 10.9402,
      "step": 9
    },
    {
      "epoch": 0.0013475071118430903,
      "eval_loss": 2.616652250289917,
      "eval_runtime": 94.4398,
      "eval_samples_per_second": 29.786,
      "eval_steps_per_second": 14.898,
      "step": 9
    },
    {
      "epoch": 0.0014972301242701003,
      "grad_norm": 11.070828437805176,
      "learning_rate": 0.0002,
      "loss": 10.0693,
      "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": 1750905435193344.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}