last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.001248335552596538,
  "eval_steps": 5,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.161118508655127e-05,
      "eval_loss": 0.9430899024009705,
      "eval_runtime": 519.9183,
      "eval_samples_per_second": 19.463,
      "eval_steps_per_second": 9.732,
      "step": 1
    },
    {
      "epoch": 0.0001248335552596538,
      "grad_norm": 0.2688457667827606,
      "learning_rate": 4e-05,
      "loss": 0.9521,
      "step": 3
    },
    {
      "epoch": 0.00020805592543275632,
      "eval_loss": 0.9386435151100159,
      "eval_runtime": 522.2398,
      "eval_samples_per_second": 19.376,
      "eval_steps_per_second": 9.689,
      "step": 5
    },
    {
      "epoch": 0.0002496671105193076,
      "grad_norm": 0.2552131712436676,
      "learning_rate": 8e-05,
      "loss": 0.9511,
      "step": 6
    },
    {
      "epoch": 0.00037450066577896136,
      "grad_norm": 0.2725978195667267,
      "learning_rate": 0.00012,
      "loss": 0.8603,
      "step": 9
    },
    {
      "epoch": 0.00041611185086551264,
      "eval_loss": 0.9117527604103088,
      "eval_runtime": 522.2012,
      "eval_samples_per_second": 19.378,
      "eval_steps_per_second": 9.69,
      "step": 10
    },
    {
      "epoch": 0.0004993342210386152,
      "grad_norm": 0.22943666577339172,
      "learning_rate": 0.00016,
      "loss": 0.8505,
      "step": 12
    },
    {
      "epoch": 0.000624167776298269,
      "grad_norm": 0.262665718793869,
      "learning_rate": 0.0002,
      "loss": 0.8551,
      "step": 15
    },
    {
      "epoch": 0.000624167776298269,
      "eval_loss": 0.892116904258728,
      "eval_runtime": 522.0076,
      "eval_samples_per_second": 19.385,
      "eval_steps_per_second": 9.693,
      "step": 15
    },
    {
      "epoch": 0.0007490013315579227,
      "grad_norm": 0.2933429181575775,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.7884,
      "step": 18
    },
    {
      "epoch": 0.0008322237017310253,
      "eval_loss": 0.8717365860939026,
      "eval_runtime": 522.3142,
      "eval_samples_per_second": 19.373,
      "eval_steps_per_second": 9.688,
      "step": 20
    },
    {
      "epoch": 0.0008738348868175766,
      "grad_norm": 0.26551419496536255,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.9068,
      "step": 21
    },
    {
      "epoch": 0.0009986684420772304,
      "grad_norm": 0.2481399029493332,
      "learning_rate": 6.909830056250527e-05,
      "loss": 0.8418,
      "step": 24
    },
    {
      "epoch": 0.0010402796271637817,
      "eval_loss": 0.8618488907814026,
      "eval_runtime": 522.4366,
      "eval_samples_per_second": 19.369,
      "eval_steps_per_second": 9.685,
      "step": 25
    },
    {
      "epoch": 0.0011235019973368841,
      "grad_norm": 0.26211848855018616,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 0.8709,
      "step": 27
    },
    {
      "epoch": 0.001248335552596538,
      "grad_norm": 0.23953785002231598,
      "learning_rate": 0.0,
      "loss": 0.8629,
      "step": 30
    },
    {
      "epoch": 0.001248335552596538,
      "eval_loss": 0.8600496649742126,
      "eval_runtime": 522.2468,
      "eval_samples_per_second": 19.376,
      "eval_steps_per_second": 9.689,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 15,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3233015371137024.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}