last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.09950248756218906,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.009950248756218905,
      "grad_norm": 0.4413871765136719,
      "learning_rate": 2e-05,
      "loss": 0.8805,
      "step": 1
    },
    {
      "epoch": 0.009950248756218905,
      "eval_loss": 0.9275743365287781,
      "eval_runtime": 5.3689,
      "eval_samples_per_second": 31.664,
      "eval_steps_per_second": 4.098,
      "step": 1
    },
    {
      "epoch": 0.01990049751243781,
      "grad_norm": 0.3896320164203644,
      "learning_rate": 4e-05,
      "loss": 0.91,
      "step": 2
    },
    {
      "epoch": 0.029850746268656716,
      "grad_norm": 0.4663248360157013,
      "learning_rate": 6e-05,
      "loss": 0.8901,
      "step": 3
    },
    {
      "epoch": 0.029850746268656716,
      "eval_loss": 0.9198755621910095,
      "eval_runtime": 5.3489,
      "eval_samples_per_second": 31.783,
      "eval_steps_per_second": 4.113,
      "step": 3
    },
    {
      "epoch": 0.03980099502487562,
      "grad_norm": 0.42104655504226685,
      "learning_rate": 8e-05,
      "loss": 0.8905,
      "step": 4
    },
    {
      "epoch": 0.04975124378109453,
      "grad_norm": 0.430433988571167,
      "learning_rate": 0.0001,
      "loss": 0.8928,
      "step": 5
    },
    {
      "epoch": 0.05970149253731343,
      "grad_norm": 0.3232385814189911,
      "learning_rate": 0.00012,
      "loss": 0.8803,
      "step": 6
    },
    {
      "epoch": 0.05970149253731343,
      "eval_loss": 0.858666181564331,
      "eval_runtime": 5.3313,
      "eval_samples_per_second": 31.887,
      "eval_steps_per_second": 4.127,
      "step": 6
    },
    {
      "epoch": 0.06965174129353234,
      "grad_norm": 0.33226680755615234,
      "learning_rate": 0.00014,
      "loss": 0.8574,
      "step": 7
    },
    {
      "epoch": 0.07960199004975124,
      "grad_norm": 0.32943281531333923,
      "learning_rate": 0.00016,
      "loss": 0.8073,
      "step": 8
    },
    {
      "epoch": 0.08955223880597014,
      "grad_norm": 0.38977158069610596,
      "learning_rate": 0.00018,
      "loss": 0.8108,
      "step": 9
    },
    {
      "epoch": 0.08955223880597014,
      "eval_loss": 0.8123431205749512,
      "eval_runtime": 5.33,
      "eval_samples_per_second": 31.895,
      "eval_steps_per_second": 4.128,
      "step": 9
    },
    {
      "epoch": 0.09950248756218906,
      "grad_norm": 0.4322220981121063,
      "learning_rate": 0.0002,
      "loss": 0.8347,
      "step": 10
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.235874343387136e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}