trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9795918367346939,
  "eval_steps": 500,
  "global_step": 64,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.030612244897959183,
      "grad_norm": 0.15814182022603535,
      "learning_rate": 1.4285714285714286e-06,
      "loss": 0.5382,
      "step": 1
    },
    {
      "epoch": 0.15306122448979592,
      "grad_norm": 0.11109193751691053,
      "learning_rate": 7.1428571428571436e-06,
      "loss": 0.4711,
      "step": 5
    },
    {
      "epoch": 0.30612244897959184,
      "grad_norm": 0.1302926162054449,
      "learning_rate": 9.931806517013612e-06,
      "loss": 0.4916,
      "step": 10
    },
    {
      "epoch": 0.45918367346938777,
      "grad_norm": 0.10174245727470003,
      "learning_rate": 9.521785803487888e-06,
      "loss": 0.4915,
      "step": 15
    },
    {
      "epoch": 0.6122448979591837,
      "grad_norm": 0.08051291607400397,
      "learning_rate": 8.770533048884483e-06,
      "loss": 0.4412,
      "step": 20
    },
    {
      "epoch": 0.7653061224489796,
      "grad_norm": 0.06459957952733913,
      "learning_rate": 7.734740790612137e-06,
      "loss": 0.4093,
      "step": 25
    },
    {
      "epoch": 0.9183673469387755,
      "grad_norm": 0.06495492947768569,
      "learning_rate": 6.492574055008474e-06,
      "loss": 0.4072,
      "step": 30
    },
    {
      "epoch": 0.9795918367346939,
      "eval_loss": 0.4253614544868469,
      "eval_runtime": 28.6513,
      "eval_samples_per_second": 19.022,
      "eval_steps_per_second": 4.782,
      "step": 32
    },
    {
      "epoch": 1.0918367346938775,
      "grad_norm": 0.06670106927688929,
      "learning_rate": 5.137771711840811e-06,
      "loss": 0.4909,
      "step": 35
    },
    {
      "epoch": 1.2448979591836735,
      "grad_norm": 0.05493919655511766,
      "learning_rate": 3.7725725642960047e-06,
      "loss": 0.3938,
      "step": 40
    },
    {
      "epoch": 1.3979591836734695,
      "grad_norm": 0.05288266918853089,
      "learning_rate": 2.5000000000000015e-06,
      "loss": 0.3768,
      "step": 45
    },
    {
      "epoch": 1.5510204081632653,
      "grad_norm": 0.05736138790419544,
      "learning_rate": 1.4160874341577447e-06,
      "loss": 0.3614,
      "step": 50
    },
    {
      "epoch": 1.7040816326530612,
      "grad_norm": 0.05167138711764061,
      "learning_rate": 6.026312439675553e-07,
      "loss": 0.3541,
      "step": 55
    },
    {
      "epoch": 1.8571428571428572,
      "grad_norm": 0.051634690642673106,
      "learning_rate": 1.210180868628219e-07,
      "loss": 0.3284,
      "step": 60
    },
    {
      "epoch": 1.9795918367346939,
      "eval_loss": 0.4213503897190094,
      "eval_runtime": 28.5952,
      "eval_samples_per_second": 19.059,
      "eval_steps_per_second": 4.791,
      "step": 64
    },
    {
      "epoch": 1.9795918367346939,
      "step": 64,
      "total_flos": 1.625730220944261e+17,
      "train_loss": 0.41580103896558285,
      "train_runtime": 940.1857,
      "train_samples_per_second": 3.333,
      "train_steps_per_second": 0.068
    }
  ],
  "logging_steps": 5,
  "max_steps": 64,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.625730220944261e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}