last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0009485866059571238,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 9.485866059571239e-05,
      "grad_norm": 0.4014851450920105,
      "learning_rate": 2e-05,
      "loss": 2.2518,
      "step": 1
    },
    {
      "epoch": 9.485866059571239e-05,
      "eval_loss": 2.141073703765869,
      "eval_runtime": 522.3107,
      "eval_samples_per_second": 8.499,
      "eval_steps_per_second": 4.25,
      "step": 1
    },
    {
      "epoch": 0.00018971732119142478,
      "grad_norm": 0.3192332088947296,
      "learning_rate": 4e-05,
      "loss": 2.2594,
      "step": 2
    },
    {
      "epoch": 0.0002845759817871372,
      "grad_norm": 0.30699437856674194,
      "learning_rate": 6e-05,
      "loss": 2.0328,
      "step": 3
    },
    {
      "epoch": 0.0002845759817871372,
      "eval_loss": 2.140320062637329,
      "eval_runtime": 521.1162,
      "eval_samples_per_second": 8.518,
      "eval_steps_per_second": 4.26,
      "step": 3
    },
    {
      "epoch": 0.00037943464238284955,
      "grad_norm": 0.3092193603515625,
      "learning_rate": 8e-05,
      "loss": 2.1988,
      "step": 4
    },
    {
      "epoch": 0.0004742933029785619,
      "grad_norm": 0.33516645431518555,
      "learning_rate": 0.0001,
      "loss": 2.2579,
      "step": 5
    },
    {
      "epoch": 0.0005691519635742744,
      "grad_norm": 0.3587290346622467,
      "learning_rate": 0.00012,
      "loss": 1.9969,
      "step": 6
    },
    {
      "epoch": 0.0005691519635742744,
      "eval_loss": 2.1289360523223877,
      "eval_runtime": 521.0815,
      "eval_samples_per_second": 8.519,
      "eval_steps_per_second": 4.26,
      "step": 6
    },
    {
      "epoch": 0.0006640106241699867,
      "grad_norm": 0.4054247736930847,
      "learning_rate": 0.00014,
      "loss": 1.9677,
      "step": 7
    },
    {
      "epoch": 0.0007588692847656991,
      "grad_norm": 0.4172271490097046,
      "learning_rate": 0.00016,
      "loss": 2.1158,
      "step": 8
    },
    {
      "epoch": 0.0008537279453614115,
      "grad_norm": 0.4382270872592926,
      "learning_rate": 0.00018,
      "loss": 2.0547,
      "step": 9
    },
    {
      "epoch": 0.0008537279453614115,
      "eval_loss": 2.0886852741241455,
      "eval_runtime": 521.2508,
      "eval_samples_per_second": 8.516,
      "eval_steps_per_second": 4.259,
      "step": 9
    },
    {
      "epoch": 0.0009485866059571238,
      "grad_norm": 0.48336905241012573,
      "learning_rate": 0.0002,
      "loss": 1.9165,
      "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": 3699128496291840.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}