last-checkpoint/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 1080,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.046296296296296294,
      "grad_norm": 5.875,
      "learning_rate": 2.2685185185185187e-05,
      "loss": 0.2551,
      "step": 50
    },
    {
      "epoch": 0.09259259259259259,
      "grad_norm": 6.84375,
      "learning_rate": 4.5833333333333334e-05,
      "loss": 0.1353,
      "step": 100
    },
    {
      "epoch": 0.1388888888888889,
      "grad_norm": 5.375,
      "learning_rate": 4.7890946502057616e-05,
      "loss": 0.1541,
      "step": 150
    },
    {
      "epoch": 0.18518518518518517,
      "grad_norm": 4.75,
      "learning_rate": 4.531893004115226e-05,
      "loss": 0.1499,
      "step": 200
    },
    {
      "epoch": 0.23148148148148148,
      "grad_norm": 4.21875,
      "learning_rate": 4.2746913580246915e-05,
      "loss": 0.1335,
      "step": 250
    },
    {
      "epoch": 0.2777777777777778,
      "grad_norm": 4.59375,
      "learning_rate": 4.017489711934157e-05,
      "loss": 0.1225,
      "step": 300
    },
    {
      "epoch": 0.32407407407407407,
      "grad_norm": 3.734375,
      "learning_rate": 3.760288065843622e-05,
      "loss": 0.1248,
      "step": 350
    },
    {
      "epoch": 0.37037037037037035,
      "grad_norm": 3.09375,
      "learning_rate": 3.503086419753087e-05,
      "loss": 0.1141,
      "step": 400
    },
    {
      "epoch": 0.4166666666666667,
      "grad_norm": 3.609375,
      "learning_rate": 3.245884773662551e-05,
      "loss": 0.1124,
      "step": 450
    },
    {
      "epoch": 0.46296296296296297,
      "grad_norm": 3.421875,
      "learning_rate": 2.9886831275720166e-05,
      "loss": 0.0966,
      "step": 500
    },
    {
      "epoch": 0.5092592592592593,
      "grad_norm": 3.21875,
      "learning_rate": 2.7314814814814816e-05,
      "loss": 0.0942,
      "step": 550
    },
    {
      "epoch": 0.5555555555555556,
      "grad_norm": 3.671875,
      "learning_rate": 2.4742798353909465e-05,
      "loss": 0.0866,
      "step": 600
    },
    {
      "epoch": 0.6018518518518519,
      "grad_norm": 2.59375,
      "learning_rate": 2.2170781893004118e-05,
      "loss": 0.0911,
      "step": 650
    },
    {
      "epoch": 0.6481481481481481,
      "grad_norm": 6.625,
      "learning_rate": 1.9598765432098768e-05,
      "loss": 0.0872,
      "step": 700
    },
    {
      "epoch": 0.6944444444444444,
      "grad_norm": 3.90625,
      "learning_rate": 1.7026748971193414e-05,
      "loss": 0.087,
      "step": 750
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": 2.34375,
      "learning_rate": 1.4454732510288065e-05,
      "loss": 0.0771,
      "step": 800
    },
    {
      "epoch": 0.7870370370370371,
      "grad_norm": 2.734375,
      "learning_rate": 1.1882716049382716e-05,
      "loss": 0.0808,
      "step": 850
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 2.40625,
      "learning_rate": 9.310699588477368e-06,
      "loss": 0.0716,
      "step": 900
    },
    {
      "epoch": 0.8796296296296297,
      "grad_norm": 2.328125,
      "learning_rate": 6.7386831275720164e-06,
      "loss": 0.0806,
      "step": 950
    },
    {
      "epoch": 0.9259259259259259,
      "grad_norm": 2.640625,
      "learning_rate": 4.166666666666667e-06,
      "loss": 0.0751,
      "step": 1000
    },
    {
      "epoch": 0.9722222222222222,
      "grad_norm": 3.734375,
      "learning_rate": 1.5946502057613169e-06,
      "loss": 0.0752,
      "step": 1050
    }
  ],
  "logging_steps": 50,
  "max_steps": 1080,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 512,
  "trial_name": null,
  "trial_params": null
}