last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00023108831039781854,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.3108831039781854e-05,
      "grad_norm": 0.3176736533641815,
      "learning_rate": 2e-05,
      "loss": 2.3064,
      "step": 1
    },
    {
      "epoch": 2.3108831039781854e-05,
      "eval_loss": 2.764178991317749,
      "eval_runtime": 184.6756,
      "eval_samples_per_second": 98.665,
      "eval_steps_per_second": 49.335,
      "step": 1
    },
    {
      "epoch": 4.621766207956371e-05,
      "grad_norm": 0.24799887835979462,
      "learning_rate": 4e-05,
      "loss": 2.6087,
      "step": 2
    },
    {
      "epoch": 6.932649311934556e-05,
      "grad_norm": 0.2642222046852112,
      "learning_rate": 6e-05,
      "loss": 2.779,
      "step": 3
    },
    {
      "epoch": 6.932649311934556e-05,
      "eval_loss": 2.764132022857666,
      "eval_runtime": 202.2714,
      "eval_samples_per_second": 90.082,
      "eval_steps_per_second": 45.043,
      "step": 3
    },
    {
      "epoch": 9.243532415912742e-05,
      "grad_norm": 0.2646380662918091,
      "learning_rate": 8e-05,
      "loss": 2.3385,
      "step": 4
    },
    {
      "epoch": 0.00011554415519890927,
      "grad_norm": 0.24436058104038239,
      "learning_rate": 0.0001,
      "loss": 2.6531,
      "step": 5
    },
    {
      "epoch": 0.00013865298623869112,
      "grad_norm": 0.24375653266906738,
      "learning_rate": 0.00012,
      "loss": 2.5198,
      "step": 6
    },
    {
      "epoch": 0.00013865298623869112,
      "eval_loss": 2.763354539871216,
      "eval_runtime": 265.9567,
      "eval_samples_per_second": 68.511,
      "eval_steps_per_second": 34.257,
      "step": 6
    },
    {
      "epoch": 0.00016176181727847296,
      "grad_norm": 0.2417832463979721,
      "learning_rate": 0.00014,
      "loss": 3.0405,
      "step": 7
    },
    {
      "epoch": 0.00018487064831825483,
      "grad_norm": 0.2619391977787018,
      "learning_rate": 0.00016,
      "loss": 3.0379,
      "step": 8
    },
    {
      "epoch": 0.00020797947935803667,
      "grad_norm": 0.2710992395877838,
      "learning_rate": 0.00018,
      "loss": 2.6747,
      "step": 9
    },
    {
      "epoch": 0.00020797947935803667,
      "eval_loss": 2.761340856552124,
      "eval_runtime": 265.6473,
      "eval_samples_per_second": 68.591,
      "eval_steps_per_second": 34.297,
      "step": 9
    },
    {
      "epoch": 0.00023108831039781854,
      "grad_norm": 0.24306856095790863,
      "learning_rate": 0.0002,
      "loss": 3.0043,
      "step": 10
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 52296291975168.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}