last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0334075723830735,
  "eval_steps": 8,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0011135857461024498,
      "eval_loss": 3.346818685531616,
      "eval_runtime": 45.6697,
      "eval_samples_per_second": 6.219,
      "eval_steps_per_second": 3.109,
      "step": 1
    },
    {
      "epoch": 0.0033407572383073497,
      "grad_norm": 3.7845871448516846,
      "learning_rate": 4e-05,
      "loss": 10.549,
      "step": 3
    },
    {
      "epoch": 0.0066815144766146995,
      "grad_norm": 4.576778888702393,
      "learning_rate": 8e-05,
      "loss": 9.166,
      "step": 6
    },
    {
      "epoch": 0.008908685968819599,
      "eval_loss": 2.684901475906372,
      "eval_runtime": 46.181,
      "eval_samples_per_second": 6.15,
      "eval_steps_per_second": 3.075,
      "step": 8
    },
    {
      "epoch": 0.01002227171492205,
      "grad_norm": 6.926581382751465,
      "learning_rate": 0.00012,
      "loss": 9.4334,
      "step": 9
    },
    {
      "epoch": 0.013363028953229399,
      "grad_norm": 4.319501876831055,
      "learning_rate": 0.00016,
      "loss": 6.3066,
      "step": 12
    },
    {
      "epoch": 0.01670378619153675,
      "grad_norm": 3.559156894683838,
      "learning_rate": 0.0002,
      "loss": 4.2514,
      "step": 15
    },
    {
      "epoch": 0.017817371937639197,
      "eval_loss": 1.1712861061096191,
      "eval_runtime": 46.2407,
      "eval_samples_per_second": 6.142,
      "eval_steps_per_second": 3.071,
      "step": 16
    },
    {
      "epoch": 0.0200445434298441,
      "grad_norm": 3.744047164916992,
      "learning_rate": 0.00018090169943749476,
      "loss": 3.8414,
      "step": 18
    },
    {
      "epoch": 0.02338530066815145,
      "grad_norm": 3.7839925289154053,
      "learning_rate": 0.00013090169943749476,
      "loss": 3.123,
      "step": 21
    },
    {
      "epoch": 0.026726057906458798,
      "grad_norm": 6.720678806304932,
      "learning_rate": 6.909830056250527e-05,
      "loss": 2.907,
      "step": 24
    },
    {
      "epoch": 0.026726057906458798,
      "eval_loss": 0.8629888296127319,
      "eval_runtime": 46.2538,
      "eval_samples_per_second": 6.14,
      "eval_steps_per_second": 3.07,
      "step": 24
    },
    {
      "epoch": 0.030066815144766147,
      "grad_norm": 3.833301305770874,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 2.6673,
      "step": 27
    },
    {
      "epoch": 0.0334075723830735,
      "grad_norm": 3.1711502075195312,
      "learning_rate": 0.0,
      "loss": 2.6537,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 15,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.42826174152704e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}