trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.6896551724137931,
  "eval_steps": 21,
  "global_step": 147,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.011494252873563218,
      "grad_norm": 0.14453125,
      "learning_rate": 0.0005,
      "loss": 0.6735,
      "step": 1
    },
    {
      "epoch": 0.2413793103448276,
      "grad_norm": 0.25390625,
      "learning_rate": 0.0005,
      "loss": 0.6075,
      "step": 21
    },
    {
      "epoch": 0.2413793103448276,
      "eval_loss": 0.5944878458976746,
      "eval_runtime": 243.3864,
      "eval_samples_per_second": 0.333,
      "eval_steps_per_second": 0.333,
      "step": 21
    },
    {
      "epoch": 0.4827586206896552,
      "grad_norm": 0.26953125,
      "learning_rate": 0.0005,
      "loss": 0.5843,
      "step": 42
    },
    {
      "epoch": 0.4827586206896552,
      "eval_loss": 0.5731382369995117,
      "eval_runtime": 242.7763,
      "eval_samples_per_second": 0.334,
      "eval_steps_per_second": 0.334,
      "step": 42
    },
    {
      "epoch": 0.7241379310344828,
      "grad_norm": 0.25390625,
      "learning_rate": 0.0005,
      "loss": 0.5643,
      "step": 63
    },
    {
      "epoch": 0.7241379310344828,
      "eval_loss": 0.5693779587745667,
      "eval_runtime": 242.4241,
      "eval_samples_per_second": 0.334,
      "eval_steps_per_second": 0.334,
      "step": 63
    },
    {
      "epoch": 0.9655172413793104,
      "grad_norm": 0.267578125,
      "learning_rate": 0.0005,
      "loss": 0.5705,
      "step": 84
    },
    {
      "epoch": 0.9655172413793104,
      "eval_loss": 0.563104510307312,
      "eval_runtime": 242.5946,
      "eval_samples_per_second": 0.334,
      "eval_steps_per_second": 0.334,
      "step": 84
    },
    {
      "epoch": 1.206896551724138,
      "grad_norm": 0.28125,
      "learning_rate": 0.0005,
      "loss": 0.4655,
      "step": 105
    },
    {
      "epoch": 1.206896551724138,
      "eval_loss": 0.5802934169769287,
      "eval_runtime": 242.4816,
      "eval_samples_per_second": 0.334,
      "eval_steps_per_second": 0.334,
      "step": 105
    },
    {
      "epoch": 1.4482758620689655,
      "grad_norm": 0.298828125,
      "learning_rate": 0.0005,
      "loss": 0.4694,
      "step": 126
    },
    {
      "epoch": 1.4482758620689655,
      "eval_loss": 0.5805184841156006,
      "eval_runtime": 242.4018,
      "eval_samples_per_second": 0.334,
      "eval_steps_per_second": 0.334,
      "step": 126
    },
    {
      "epoch": 1.6896551724137931,
      "grad_norm": 0.330078125,
      "learning_rate": 0.0005,
      "loss": 0.4568,
      "step": 147
    },
    {
      "epoch": 1.6896551724137931,
      "eval_loss": 0.5760032534599304,
      "eval_runtime": 242.3367,
      "eval_samples_per_second": 0.334,
      "eval_steps_per_second": 0.334,
      "step": 147
    }
  ],
  "logging_steps": 21,
  "max_steps": 261,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 21,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6.46118484598825e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}