last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0016300582745833164,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00016300582745833165,
      "grad_norm": 1.1928633451461792,
      "learning_rate": 2e-05,
      "loss": 1.4233,
      "step": 1
    },
    {
      "epoch": 0.00016300582745833165,
      "eval_loss": 1.4654473066329956,
      "eval_runtime": 1317.8508,
      "eval_samples_per_second": 1.96,
      "eval_steps_per_second": 0.98,
      "step": 1
    },
    {
      "epoch": 0.0003260116549166633,
      "grad_norm": 0.952515184879303,
      "learning_rate": 4e-05,
      "loss": 1.5684,
      "step": 2
    },
    {
      "epoch": 0.0004890174823749949,
      "grad_norm": 1.6886141300201416,
      "learning_rate": 6e-05,
      "loss": 1.5758,
      "step": 3
    },
    {
      "epoch": 0.0004890174823749949,
      "eval_loss": 1.461408257484436,
      "eval_runtime": 1320.497,
      "eval_samples_per_second": 1.956,
      "eval_steps_per_second": 0.978,
      "step": 3
    },
    {
      "epoch": 0.0006520233098333266,
      "grad_norm": 1.8855215311050415,
      "learning_rate": 8e-05,
      "loss": 1.9548,
      "step": 4
    },
    {
      "epoch": 0.0008150291372916582,
      "grad_norm": 1.4256668090820312,
      "learning_rate": 0.0001,
      "loss": 1.3427,
      "step": 5
    },
    {
      "epoch": 0.0009780349647499898,
      "grad_norm": 0.9615749716758728,
      "learning_rate": 0.00012,
      "loss": 1.5318,
      "step": 6
    },
    {
      "epoch": 0.0009780349647499898,
      "eval_loss": 1.399588704109192,
      "eval_runtime": 1318.9507,
      "eval_samples_per_second": 1.958,
      "eval_steps_per_second": 0.98,
      "step": 6
    },
    {
      "epoch": 0.0011410407922083214,
      "grad_norm": 0.9091509580612183,
      "learning_rate": 0.00014,
      "loss": 1.4637,
      "step": 7
    },
    {
      "epoch": 0.0013040466196666532,
      "grad_norm": 4.046209335327148,
      "learning_rate": 0.00016,
      "loss": 1.9125,
      "step": 8
    },
    {
      "epoch": 0.0014670524471249848,
      "grad_norm": 0.7425318956375122,
      "learning_rate": 0.00018,
      "loss": 1.3163,
      "step": 9
    },
    {
      "epoch": 0.0014670524471249848,
      "eval_loss": 1.2187490463256836,
      "eval_runtime": 1319.7023,
      "eval_samples_per_second": 1.957,
      "eval_steps_per_second": 0.979,
      "step": 9
    },
    {
      "epoch": 0.0016300582745833164,
      "grad_norm": 0.7953963875770569,
      "learning_rate": 0.0002,
      "loss": 1.0503,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.483774567120896e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}