last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0001149410607782787,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.2771228975364299e-05,
      "grad_norm": 1.0573991537094116,
      "learning_rate": 2e-05,
      "loss": 3.2321,
      "step": 1
    },
    {
      "epoch": 1.2771228975364299e-05,
      "eval_loss": 3.49228572845459,
      "eval_runtime": 302.7222,
      "eval_samples_per_second": 108.908,
      "eval_steps_per_second": 54.456,
      "step": 1
    },
    {
      "epoch": 2.5542457950728598e-05,
      "grad_norm": 1.0892497301101685,
      "learning_rate": 4e-05,
      "loss": 3.4583,
      "step": 2
    },
    {
      "epoch": 3.8313686926092896e-05,
      "grad_norm": 1.0131731033325195,
      "learning_rate": 6e-05,
      "loss": 3.43,
      "step": 3
    },
    {
      "epoch": 3.8313686926092896e-05,
      "eval_loss": 3.4900827407836914,
      "eval_runtime": 303.0453,
      "eval_samples_per_second": 108.792,
      "eval_steps_per_second": 54.398,
      "step": 3
    },
    {
      "epoch": 5.1084915901457196e-05,
      "grad_norm": 0.96201092004776,
      "learning_rate": 8e-05,
      "loss": 3.4506,
      "step": 4
    },
    {
      "epoch": 6.38561448768215e-05,
      "grad_norm": 1.1099013090133667,
      "learning_rate": 0.0001,
      "loss": 3.3963,
      "step": 5
    },
    {
      "epoch": 7.662737385218579e-05,
      "grad_norm": 1.070656418800354,
      "learning_rate": 0.00012,
      "loss": 3.2797,
      "step": 6
    },
    {
      "epoch": 7.662737385218579e-05,
      "eval_loss": 3.453162670135498,
      "eval_runtime": 302.0746,
      "eval_samples_per_second": 109.142,
      "eval_steps_per_second": 54.573,
      "step": 6
    },
    {
      "epoch": 8.93986028275501e-05,
      "grad_norm": 1.0605616569519043,
      "learning_rate": 0.00014,
      "loss": 3.6806,
      "step": 7
    },
    {
      "epoch": 0.00010216983180291439,
      "grad_norm": 1.1009156703948975,
      "learning_rate": 0.00016,
      "loss": 3.5007,
      "step": 8
    },
    {
      "epoch": 0.0001149410607782787,
      "grad_norm": 1.0189341306686401,
      "learning_rate": 0.00018,
      "loss": 3.4677,
      "step": 9
    },
    {
      "epoch": 0.0001149410607782787,
      "eval_loss": 3.3343887329101562,
      "eval_runtime": 304.9849,
      "eval_samples_per_second": 108.1,
      "eval_steps_per_second": 54.052,
      "step": 9
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 30863385427968.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}