last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004611246317407455,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005123607019341616,
      "grad_norm": 0.3183790445327759,
      "learning_rate": 2e-05,
      "loss": 0.7016,
      "step": 1
    },
    {
      "epoch": 0.0005123607019341616,
      "eval_loss": 1.1740705966949463,
      "eval_runtime": 91.1727,
      "eval_samples_per_second": 4.508,
      "eval_steps_per_second": 4.508,
      "step": 1
    },
    {
      "epoch": 0.0010247214038683233,
      "grad_norm": 0.9751269221305847,
      "learning_rate": 4e-05,
      "loss": 0.6813,
      "step": 2
    },
    {
      "epoch": 0.001537082105802485,
      "grad_norm": 0.8691899180412292,
      "learning_rate": 6e-05,
      "loss": 0.6713,
      "step": 3
    },
    {
      "epoch": 0.001537082105802485,
      "eval_loss": 1.1680519580841064,
      "eval_runtime": 89.7416,
      "eval_samples_per_second": 4.58,
      "eval_steps_per_second": 4.58,
      "step": 3
    },
    {
      "epoch": 0.0020494428077366465,
      "grad_norm": 0.8441597819328308,
      "learning_rate": 8e-05,
      "loss": 0.9886,
      "step": 4
    },
    {
      "epoch": 0.0025618035096708082,
      "grad_norm": 0.9734748005867004,
      "learning_rate": 0.0001,
      "loss": 0.9814,
      "step": 5
    },
    {
      "epoch": 0.00307416421160497,
      "grad_norm": 0.8902590870857239,
      "learning_rate": 0.00012,
      "loss": 0.7274,
      "step": 6
    },
    {
      "epoch": 0.00307416421160497,
      "eval_loss": 1.1203179359436035,
      "eval_runtime": 89.7709,
      "eval_samples_per_second": 4.578,
      "eval_steps_per_second": 4.578,
      "step": 6
    },
    {
      "epoch": 0.0035865249135391317,
      "grad_norm": 0.5923784375190735,
      "learning_rate": 0.00014,
      "loss": 1.2382,
      "step": 7
    },
    {
      "epoch": 0.004098885615473293,
      "grad_norm": 0.6953917145729065,
      "learning_rate": 0.00016,
      "loss": 1.271,
      "step": 8
    },
    {
      "epoch": 0.004611246317407455,
      "grad_norm": 0.8029590249061584,
      "learning_rate": 0.00018,
      "loss": 0.6573,
      "step": 9
    },
    {
      "epoch": 0.004611246317407455,
      "eval_loss": 1.064617395401001,
      "eval_runtime": 89.7935,
      "eval_samples_per_second": 4.577,
      "eval_steps_per_second": 4.577,
      "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": 6291348239941632.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}