last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.025660215973484442,
  "eval_steps": 5,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0017106810648989629,
      "grad_norm": 0.5461569428443909,
      "learning_rate": 1e-05,
      "loss": 3.6187,
      "step": 1
    },
    {
      "epoch": 0.0017106810648989629,
      "eval_loss": 3.4791719913482666,
      "eval_runtime": 44.1144,
      "eval_samples_per_second": 22.328,
      "eval_steps_per_second": 11.176,
      "step": 1
    },
    {
      "epoch": 0.0034213621297979257,
      "grad_norm": 0.5224133729934692,
      "learning_rate": 2e-05,
      "loss": 3.4447,
      "step": 2
    },
    {
      "epoch": 0.005132043194696889,
      "grad_norm": 0.600152313709259,
      "learning_rate": 3e-05,
      "loss": 3.4106,
      "step": 3
    },
    {
      "epoch": 0.0068427242595958514,
      "grad_norm": 0.5289663672447205,
      "learning_rate": 4e-05,
      "loss": 3.551,
      "step": 4
    },
    {
      "epoch": 0.008553405324494815,
      "grad_norm": 0.5006380081176758,
      "learning_rate": 5e-05,
      "loss": 3.3413,
      "step": 5
    },
    {
      "epoch": 0.008553405324494815,
      "eval_loss": 3.4636945724487305,
      "eval_runtime": 3.823,
      "eval_samples_per_second": 257.654,
      "eval_steps_per_second": 128.958,
      "step": 5
    },
    {
      "epoch": 0.010264086389393778,
      "grad_norm": 0.4884447753429413,
      "learning_rate": 6e-05,
      "loss": 3.3282,
      "step": 6
    },
    {
      "epoch": 0.01197476745429274,
      "grad_norm": 0.5128330588340759,
      "learning_rate": 7e-05,
      "loss": 3.3899,
      "step": 7
    },
    {
      "epoch": 0.013685448519191703,
      "grad_norm": 0.4853476583957672,
      "learning_rate": 8e-05,
      "loss": 3.4673,
      "step": 8
    },
    {
      "epoch": 0.015396129584090666,
      "grad_norm": 0.4155305325984955,
      "learning_rate": 9e-05,
      "loss": 3.1119,
      "step": 9
    },
    {
      "epoch": 0.01710681064898963,
      "grad_norm": 0.5410354733467102,
      "learning_rate": 0.0001,
      "loss": 3.4547,
      "step": 10
    },
    {
      "epoch": 0.01710681064898963,
      "eval_loss": 3.410022020339966,
      "eval_runtime": 3.9919,
      "eval_samples_per_second": 246.751,
      "eval_steps_per_second": 123.501,
      "step": 10
    },
    {
      "epoch": 0.018817491713888593,
      "grad_norm": 0.4462915062904358,
      "learning_rate": 9.755282581475769e-05,
      "loss": 3.4584,
      "step": 11
    },
    {
      "epoch": 0.020528172778787556,
      "grad_norm": 0.48912936449050903,
      "learning_rate": 9.045084971874738e-05,
      "loss": 3.4649,
      "step": 12
    },
    {
      "epoch": 0.02223885384368652,
      "grad_norm": 0.5210543870925903,
      "learning_rate": 7.938926261462366e-05,
      "loss": 3.4684,
      "step": 13
    },
    {
      "epoch": 0.02394953490858548,
      "grad_norm": 0.4840293228626251,
      "learning_rate": 6.545084971874738e-05,
      "loss": 3.4167,
      "step": 14
    },
    {
      "epoch": 0.025660215973484442,
      "grad_norm": 0.4712510406970978,
      "learning_rate": 5e-05,
      "loss": 3.3471,
      "step": 15
    },
    {
      "epoch": 0.025660215973484442,
      "eval_loss": 3.337122678756714,
      "eval_runtime": 3.8738,
      "eval_samples_per_second": 254.275,
      "eval_steps_per_second": 127.267,
      "step": 15
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 205755902853120.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}