last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.028070175438596492,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002807017543859649,
      "grad_norm": 0.36724358797073364,
      "learning_rate": 2e-05,
      "loss": 2.4026,
      "step": 1
    },
    {
      "epoch": 0.002807017543859649,
      "eval_loss": 2.408853054046631,
      "eval_runtime": 4.9311,
      "eval_samples_per_second": 15.21,
      "eval_steps_per_second": 15.21,
      "step": 1
    },
    {
      "epoch": 0.005614035087719298,
      "grad_norm": 0.3240700364112854,
      "learning_rate": 4e-05,
      "loss": 2.1548,
      "step": 2
    },
    {
      "epoch": 0.008421052631578947,
      "grad_norm": 0.41011378169059753,
      "learning_rate": 6e-05,
      "loss": 2.3262,
      "step": 3
    },
    {
      "epoch": 0.008421052631578947,
      "eval_loss": 2.407918691635132,
      "eval_runtime": 4.9532,
      "eval_samples_per_second": 15.142,
      "eval_steps_per_second": 15.142,
      "step": 3
    },
    {
      "epoch": 0.011228070175438596,
      "grad_norm": 0.38444051146507263,
      "learning_rate": 8e-05,
      "loss": 2.2965,
      "step": 4
    },
    {
      "epoch": 0.014035087719298246,
      "grad_norm": 0.3640546500682831,
      "learning_rate": 0.0001,
      "loss": 2.2651,
      "step": 5
    },
    {
      "epoch": 0.016842105263157894,
      "grad_norm": 0.4015711545944214,
      "learning_rate": 0.00012,
      "loss": 2.3536,
      "step": 6
    },
    {
      "epoch": 0.016842105263157894,
      "eval_loss": 2.4016966819763184,
      "eval_runtime": 5.1087,
      "eval_samples_per_second": 14.681,
      "eval_steps_per_second": 14.681,
      "step": 6
    },
    {
      "epoch": 0.019649122807017545,
      "grad_norm": 0.33648228645324707,
      "learning_rate": 0.00014,
      "loss": 2.3676,
      "step": 7
    },
    {
      "epoch": 0.02245614035087719,
      "grad_norm": 0.3176080584526062,
      "learning_rate": 0.00016,
      "loss": 2.4582,
      "step": 8
    },
    {
      "epoch": 0.02526315789473684,
      "grad_norm": 0.3221423625946045,
      "learning_rate": 0.00018,
      "loss": 2.1804,
      "step": 9
    },
    {
      "epoch": 0.02526315789473684,
      "eval_loss": 2.385359764099121,
      "eval_runtime": 5.0175,
      "eval_samples_per_second": 14.948,
      "eval_steps_per_second": 14.948,
      "step": 9
    },
    {
      "epoch": 0.028070175438596492,
      "grad_norm": 0.395561158657074,
      "learning_rate": 0.0002,
      "loss": 2.4822,
      "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": 1628733135912960.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}