trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9392265193370166,
  "eval_steps": 22,
  "global_step": 176,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.011049723756906077,
      "grad_norm": 0.21801863610744476,
      "learning_rate": 0.0005,
      "loss": 0.7346,
      "step": 1
    },
    {
      "epoch": 0.2430939226519337,
      "grad_norm": 0.31375670433044434,
      "learning_rate": 0.0005,
      "loss": 0.6429,
      "step": 22
    },
    {
      "epoch": 0.2430939226519337,
      "eval_loss": 0.6063512563705444,
      "eval_runtime": 264.3898,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 22
    },
    {
      "epoch": 0.4861878453038674,
      "grad_norm": 0.22838832437992096,
      "learning_rate": 0.0005,
      "loss": 0.5907,
      "step": 44
    },
    {
      "epoch": 0.4861878453038674,
      "eval_loss": 0.589008629322052,
      "eval_runtime": 264.4169,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 44
    },
    {
      "epoch": 0.7292817679558011,
      "grad_norm": 0.24711200594902039,
      "learning_rate": 0.0005,
      "loss": 0.5933,
      "step": 66
    },
    {
      "epoch": 0.7292817679558011,
      "eval_loss": 0.5814085006713867,
      "eval_runtime": 264.3062,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 66
    },
    {
      "epoch": 0.9723756906077348,
      "grad_norm": 0.2660823166370392,
      "learning_rate": 0.0005,
      "loss": 0.5547,
      "step": 88
    },
    {
      "epoch": 0.9723756906077348,
      "eval_loss": 0.573217511177063,
      "eval_runtime": 264.4531,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 88
    },
    {
      "epoch": 1.2099447513812154,
      "grad_norm": 0.3317931592464447,
      "learning_rate": 0.0005,
      "loss": 0.4669,
      "step": 110
    },
    {
      "epoch": 1.2099447513812154,
      "eval_loss": 0.5893386006355286,
      "eval_runtime": 264.4213,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 110
    },
    {
      "epoch": 1.4530386740331491,
      "grad_norm": 0.3534264862537384,
      "learning_rate": 0.0005,
      "loss": 0.4662,
      "step": 132
    },
    {
      "epoch": 1.4530386740331491,
      "eval_loss": 0.5882214307785034,
      "eval_runtime": 264.4779,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 132
    },
    {
      "epoch": 1.6961325966850829,
      "grad_norm": 0.30404219031333923,
      "learning_rate": 0.0005,
      "loss": 0.4649,
      "step": 154
    },
    {
      "epoch": 1.6961325966850829,
      "eval_loss": 0.5874143242835999,
      "eval_runtime": 264.3455,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 154
    },
    {
      "epoch": 1.9392265193370166,
      "grad_norm": 0.32281291484832764,
      "learning_rate": 0.0005,
      "loss": 0.463,
      "step": 176
    },
    {
      "epoch": 1.9392265193370166,
      "eval_loss": 0.5917199850082397,
      "eval_runtime": 264.4079,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 176
    }
  ],
  "logging_steps": 22,
  "max_steps": 270,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 22,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 7.82706388705407e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}