trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9924812030075187,
  "eval_steps": 500,
  "global_step": 66,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.015037593984962405,
      "grad_norm": 2.8635096549987793,
      "learning_rate": 2.857142857142857e-05,
      "loss": 5.568,
      "step": 1
    },
    {
      "epoch": 0.07518796992481203,
      "grad_norm": 3.604334831237793,
      "learning_rate": 0.00014285714285714287,
      "loss": 5.5365,
      "step": 5
    },
    {
      "epoch": 0.15037593984962405,
      "grad_norm": 4.168403625488281,
      "learning_rate": 0.00019872683547213446,
      "loss": 4.6649,
      "step": 10
    },
    {
      "epoch": 0.22556390977443608,
      "grad_norm": 4.2408342361450195,
      "learning_rate": 0.00019106347728549135,
      "loss": 3.5383,
      "step": 15
    },
    {
      "epoch": 0.3007518796992481,
      "grad_norm": 2.7263500690460205,
      "learning_rate": 0.00017698339834299061,
      "loss": 2.7727,
      "step": 20
    },
    {
      "epoch": 0.37593984962406013,
      "grad_norm": 1.255944848060608,
      "learning_rate": 0.0001574787410214407,
      "loss": 2.2585,
      "step": 25
    },
    {
      "epoch": 0.45112781954887216,
      "grad_norm": 0.8340353965759277,
      "learning_rate": 0.00013392388661180303,
      "loss": 1.9264,
      "step": 30
    },
    {
      "epoch": 0.5263157894736842,
      "grad_norm": 0.7235950827598572,
      "learning_rate": 0.00010797861055530831,
      "loss": 1.7245,
      "step": 35
    },
    {
      "epoch": 0.6015037593984962,
      "grad_norm": 0.505240261554718,
      "learning_rate": 8.147112759128859e-05,
      "loss": 1.5989,
      "step": 40
    },
    {
      "epoch": 0.6766917293233082,
      "grad_norm": 0.5301125049591064,
      "learning_rate": 5.626926795411447e-05,
      "loss": 1.5427,
      "step": 45
    },
    {
      "epoch": 0.7518796992481203,
      "grad_norm": 0.3930007517337799,
      "learning_rate": 3.414886209349615e-05,
      "loss": 1.5057,
      "step": 50
    },
    {
      "epoch": 0.8270676691729323,
      "grad_norm": 0.39453694224357605,
      "learning_rate": 1.6668608091748495e-05,
      "loss": 1.4899,
      "step": 55
    },
    {
      "epoch": 0.9022556390977443,
      "grad_norm": 0.38447922468185425,
      "learning_rate": 5.060239153161872e-06,
      "loss": 1.483,
      "step": 60
    },
    {
      "epoch": 0.9774436090225563,
      "grad_norm": 0.3598044216632843,
      "learning_rate": 1.4173043232380557e-07,
      "loss": 1.4849,
      "step": 65
    },
    {
      "epoch": 0.9924812030075187,
      "eval_loss": 2.0512757301330566,
      "eval_runtime": 2.2387,
      "eval_samples_per_second": 11.167,
      "eval_steps_per_second": 0.447,
      "step": 66
    },
    {
      "epoch": 0.9924812030075187,
      "step": 66,
      "total_flos": 1.3348374772438794e+18,
      "train_loss": 2.4112901976614287,
      "train_runtime": 910.693,
      "train_samples_per_second": 37.245,
      "train_steps_per_second": 0.072
    }
  ],
  "logging_steps": 5,
  "max_steps": 66,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3348374772438794e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}