trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9690721649484537,
  "eval_steps": 30,
  "global_step": 18,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16494845360824742,
      "grad_norm": 1.600954532623291,
      "learning_rate": 2.8333333333333332e-05,
      "loss": 10.0942,
      "step": 1
    },
    {
      "epoch": 0.32989690721649484,
      "grad_norm": 1.7290301322937012,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 10.2287,
      "step": 2
    },
    {
      "epoch": 0.4948453608247423,
      "grad_norm": 1.7419556379318237,
      "learning_rate": 2.5e-05,
      "loss": 10.0563,
      "step": 3
    },
    {
      "epoch": 0.6597938144329897,
      "grad_norm": 2.1707518100738525,
      "learning_rate": 2.3333333333333336e-05,
      "loss": 10.2105,
      "step": 4
    },
    {
      "epoch": 0.8247422680412371,
      "grad_norm": 2.1088027954101562,
      "learning_rate": 2.1666666666666667e-05,
      "loss": 10.0726,
      "step": 5
    },
    {
      "epoch": 0.9896907216494846,
      "grad_norm": 2.13659405708313,
      "learning_rate": 1.9999999999999998e-05,
      "loss": 9.879,
      "step": 6
    },
    {
      "epoch": 1.1546391752577319,
      "grad_norm": 2.2547099590301514,
      "learning_rate": 1.8333333333333336e-05,
      "loss": 9.9348,
      "step": 7
    },
    {
      "epoch": 1.3195876288659794,
      "grad_norm": NaN,
      "learning_rate": 1.8333333333333336e-05,
      "loss": 15.4531,
      "step": 8
    },
    {
      "epoch": 1.4845360824742269,
      "grad_norm": 2.7877187728881836,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 10.0002,
      "step": 9
    },
    {
      "epoch": 1.6494845360824741,
      "grad_norm": 2.4134161472320557,
      "learning_rate": 1.5e-05,
      "loss": 9.7697,
      "step": 10
    },
    {
      "epoch": 1.8144329896907216,
      "grad_norm": 2.528310775756836,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 9.7906,
      "step": 11
    },
    {
      "epoch": 1.9793814432989691,
      "grad_norm": 2.7709970474243164,
      "learning_rate": 1.1666666666666668e-05,
      "loss": 9.7277,
      "step": 12
    },
    {
      "epoch": 2.1443298969072164,
      "grad_norm": 2.866523027420044,
      "learning_rate": 9.999999999999999e-06,
      "loss": 9.7397,
      "step": 13
    },
    {
      "epoch": 2.3092783505154637,
      "grad_norm": 2.8438379764556885,
      "learning_rate": 8.333333333333334e-06,
      "loss": 9.641,
      "step": 14
    },
    {
      "epoch": 2.4742268041237114,
      "grad_norm": 3.0742781162261963,
      "learning_rate": 6.666666666666667e-06,
      "loss": 9.6901,
      "step": 15
    },
    {
      "epoch": 2.6391752577319587,
      "grad_norm": 3.056804895401001,
      "learning_rate": 4.9999999999999996e-06,
      "loss": 9.6277,
      "step": 16
    },
    {
      "epoch": 2.804123711340206,
      "grad_norm": 3.2071685791015625,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 9.6238,
      "step": 17
    },
    {
      "epoch": 2.9690721649484537,
      "grad_norm": 3.0884222984313965,
      "learning_rate": 1.6666666666666667e-06,
      "loss": 9.5605,
      "step": 18
    },
    {
      "epoch": 2.9690721649484537,
      "step": 18,
      "total_flos": 293443795561212.0,
      "train_loss": 10.172218322753906,
      "train_runtime": 1046.7105,
      "train_samples_per_second": 0.278,
      "train_steps_per_second": 0.017
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 18,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 293443795561212.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}