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": 1.6202701330184937,
      "learning_rate": 2.8333333333333332e-05,
      "loss": 10.1046,
      "step": 1
    },
    {
      "epoch": 0.32989690721649484,
      "grad_norm": 1.7503541707992554,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 10.2343,
      "step": 2
    },
    {
      "epoch": 0.4948453608247423,
      "grad_norm": 1.7385201454162598,
      "learning_rate": 2.5e-05,
      "loss": 10.0556,
      "step": 3
    },
    {
      "epoch": 0.6597938144329897,
      "grad_norm": 2.09757399559021,
      "learning_rate": 2.3333333333333336e-05,
      "loss": 10.1984,
      "step": 4
    },
    {
      "epoch": 0.8247422680412371,
      "grad_norm": 2.21995210647583,
      "learning_rate": 2.1666666666666667e-05,
      "loss": 10.0701,
      "step": 5
    },
    {
      "epoch": 0.9896907216494846,
      "grad_norm": 2.1417298316955566,
      "learning_rate": 1.9999999999999998e-05,
      "loss": 9.8698,
      "step": 6
    },
    {
      "epoch": 1.1546391752577319,
      "grad_norm": 2.234846591949463,
      "learning_rate": 1.8333333333333336e-05,
      "loss": 9.9373,
      "step": 7
    },
    {
      "epoch": 1.3195876288659794,
      "grad_norm": 2.584611654281616,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 9.912,
      "step": 8
    },
    {
      "epoch": 1.4845360824742269,
      "grad_norm": 2.8143157958984375,
      "learning_rate": 1.5e-05,
      "loss": 9.9623,
      "step": 9
    },
    {
      "epoch": 1.6494845360824741,
      "grad_norm": 2.415213108062744,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 9.7267,
      "step": 10
    },
    {
      "epoch": 1.8144329896907216,
      "grad_norm": 2.538069725036621,
      "learning_rate": 1.1666666666666668e-05,
      "loss": 9.7572,
      "step": 11
    },
    {
      "epoch": 1.9793814432989691,
      "grad_norm": 2.8651926517486572,
      "learning_rate": 9.999999999999999e-06,
      "loss": 9.6861,
      "step": 12
    },
    {
      "epoch": 2.1443298969072164,
      "grad_norm": 2.910115957260132,
      "learning_rate": 8.333333333333334e-06,
      "loss": 9.7232,
      "step": 13
    },
    {
      "epoch": 2.3092783505154637,
      "grad_norm": 3.0204553604125977,
      "learning_rate": 6.666666666666667e-06,
      "loss": 9.6253,
      "step": 14
    },
    {
      "epoch": 2.4742268041237114,
      "grad_norm": 3.151297092437744,
      "learning_rate": 4.9999999999999996e-06,
      "loss": 9.6825,
      "step": 15
    },
    {
      "epoch": 2.6391752577319587,
      "grad_norm": 3.1233134269714355,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 9.5969,
      "step": 16
    },
    {
      "epoch": 2.804123711340206,
      "grad_norm": 3.1847848892211914,
      "learning_rate": 1.6666666666666667e-06,
      "loss": 9.5601,
      "step": 17
    },
    {
      "epoch": 2.9690721649484537,
      "grad_norm": 3.147383451461792,
      "learning_rate": 0.0,
      "loss": 9.5912,
      "step": 18
    },
    {
      "epoch": 2.9690721649484537,
      "step": 18,
      "total_flos": 212383098180048.0,
      "train_loss": 9.84964762793647,
      "train_runtime": 1795.9951,
      "train_samples_per_second": 0.162,
      "train_steps_per_second": 0.01
    }
  ],
  "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": 212383098180048.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}