last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00041520096880226054,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.613344097802895e-05,
      "grad_norm": 0.6174605488777161,
      "learning_rate": 2e-05,
      "loss": 0.9053,
      "step": 1
    },
    {
      "epoch": 4.613344097802895e-05,
      "eval_loss": 0.9036161303520203,
      "eval_runtime": 350.8421,
      "eval_samples_per_second": 26.015,
      "eval_steps_per_second": 13.009,
      "step": 1
    },
    {
      "epoch": 9.22668819560579e-05,
      "grad_norm": 0.603773832321167,
      "learning_rate": 4e-05,
      "loss": 0.9049,
      "step": 2
    },
    {
      "epoch": 0.00013840032293408684,
      "grad_norm": 0.5677074193954468,
      "learning_rate": 6e-05,
      "loss": 0.9731,
      "step": 3
    },
    {
      "epoch": 0.00013840032293408684,
      "eval_loss": 0.9001439809799194,
      "eval_runtime": 348.8784,
      "eval_samples_per_second": 26.161,
      "eval_steps_per_second": 13.082,
      "step": 3
    },
    {
      "epoch": 0.0001845337639121158,
      "grad_norm": 0.5888237953186035,
      "learning_rate": 8e-05,
      "loss": 0.8853,
      "step": 4
    },
    {
      "epoch": 0.00023066720489014475,
      "grad_norm": 0.5911535024642944,
      "learning_rate": 0.0001,
      "loss": 0.7564,
      "step": 5
    },
    {
      "epoch": 0.0002768006458681737,
      "grad_norm": 0.6763503551483154,
      "learning_rate": 0.00012,
      "loss": 0.913,
      "step": 6
    },
    {
      "epoch": 0.0002768006458681737,
      "eval_loss": 0.8399578332901001,
      "eval_runtime": 349.1528,
      "eval_samples_per_second": 26.14,
      "eval_steps_per_second": 13.072,
      "step": 6
    },
    {
      "epoch": 0.00032293408684620263,
      "grad_norm": 0.682282567024231,
      "learning_rate": 0.00014,
      "loss": 0.8114,
      "step": 7
    },
    {
      "epoch": 0.0003690675278242316,
      "grad_norm": 0.69523686170578,
      "learning_rate": 0.00016,
      "loss": 0.8753,
      "step": 8
    },
    {
      "epoch": 0.00041520096880226054,
      "grad_norm": 0.6659558415412903,
      "learning_rate": 0.00018,
      "loss": 0.6686,
      "step": 9
    },
    {
      "epoch": 0.00041520096880226054,
      "eval_loss": 0.6748826503753662,
      "eval_runtime": 349.4345,
      "eval_samples_per_second": 26.119,
      "eval_steps_per_second": 13.061,
      "step": 9
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2034556106440704.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}