last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004429678848283499,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00044296788482834997,
      "grad_norm": 0.6514183878898621,
      "learning_rate": 2e-05,
      "loss": 5.6127,
      "step": 1
    },
    {
      "epoch": 0.00044296788482834997,
      "eval_loss": 1.6711302995681763,
      "eval_runtime": 27.0627,
      "eval_samples_per_second": 35.141,
      "eval_steps_per_second": 17.589,
      "step": 1
    },
    {
      "epoch": 0.0008859357696566999,
      "grad_norm": 0.6110041737556458,
      "learning_rate": 4e-05,
      "loss": 6.4331,
      "step": 2
    },
    {
      "epoch": 0.00132890365448505,
      "grad_norm": 0.7563453912734985,
      "learning_rate": 6e-05,
      "loss": 5.9224,
      "step": 3
    },
    {
      "epoch": 0.00132890365448505,
      "eval_loss": 1.6708110570907593,
      "eval_runtime": 25.6097,
      "eval_samples_per_second": 37.134,
      "eval_steps_per_second": 18.587,
      "step": 3
    },
    {
      "epoch": 0.0017718715393133999,
      "grad_norm": 0.8480954170227051,
      "learning_rate": 8e-05,
      "loss": 5.6718,
      "step": 4
    },
    {
      "epoch": 0.0022148394241417496,
      "grad_norm": 1.3175694942474365,
      "learning_rate": 0.0001,
      "loss": 6.6246,
      "step": 5
    },
    {
      "epoch": 0.0026578073089701,
      "grad_norm": 1.2627931833267212,
      "learning_rate": 0.00012,
      "loss": 7.2104,
      "step": 6
    },
    {
      "epoch": 0.0026578073089701,
      "eval_loss": 1.6560412645339966,
      "eval_runtime": 25.7784,
      "eval_samples_per_second": 36.891,
      "eval_steps_per_second": 18.465,
      "step": 6
    },
    {
      "epoch": 0.0031007751937984496,
      "grad_norm": 1.5461033582687378,
      "learning_rate": 0.00014,
      "loss": 6.9512,
      "step": 7
    },
    {
      "epoch": 0.0035437430786267998,
      "grad_norm": 1.9836843013763428,
      "learning_rate": 0.00016,
      "loss": 6.8547,
      "step": 8
    },
    {
      "epoch": 0.003986710963455149,
      "grad_norm": 2.3089280128479004,
      "learning_rate": 0.00018,
      "loss": 7.3323,
      "step": 9
    },
    {
      "epoch": 0.003986710963455149,
      "eval_loss": 1.5851309299468994,
      "eval_runtime": 25.7305,
      "eval_samples_per_second": 36.96,
      "eval_steps_per_second": 18.499,
      "step": 9
    },
    {
      "epoch": 0.004429678848283499,
      "grad_norm": 3.040799140930176,
      "learning_rate": 0.0002,
      "loss": 6.1683,
      "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": 1632478030725120.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}