last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0005465451514613251,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 5.465451514613251e-05,
      "grad_norm": 0.9837482571601868,
      "learning_rate": 2e-05,
      "loss": 2.145,
      "step": 1
    },
    {
      "epoch": 5.465451514613251e-05,
      "eval_loss": 2.549757957458496,
      "eval_runtime": 428.228,
      "eval_samples_per_second": 17.99,
      "eval_steps_per_second": 8.995,
      "step": 1
    },
    {
      "epoch": 0.00010930903029226501,
      "grad_norm": 0.9399062991142273,
      "learning_rate": 4e-05,
      "loss": 2.2083,
      "step": 2
    },
    {
      "epoch": 0.00016396354543839753,
      "grad_norm": 1.6526497602462769,
      "learning_rate": 6e-05,
      "loss": 2.7497,
      "step": 3
    },
    {
      "epoch": 0.00016396354543839753,
      "eval_loss": 2.5405495166778564,
      "eval_runtime": 426.8486,
      "eval_samples_per_second": 18.049,
      "eval_steps_per_second": 9.024,
      "step": 3
    },
    {
      "epoch": 0.00021861806058453003,
      "grad_norm": 0.9579558968544006,
      "learning_rate": 8e-05,
      "loss": 2.141,
      "step": 4
    },
    {
      "epoch": 0.00027327257573066255,
      "grad_norm": 1.4143273830413818,
      "learning_rate": 0.0001,
      "loss": 2.1524,
      "step": 5
    },
    {
      "epoch": 0.00032792709087679507,
      "grad_norm": 1.2946897745132446,
      "learning_rate": 0.00012,
      "loss": 1.9838,
      "step": 6
    },
    {
      "epoch": 0.00032792709087679507,
      "eval_loss": 2.34071683883667,
      "eval_runtime": 427.3927,
      "eval_samples_per_second": 18.026,
      "eval_steps_per_second": 9.013,
      "step": 6
    },
    {
      "epoch": 0.0003825816060229276,
      "grad_norm": 1.4767484664916992,
      "learning_rate": 0.00014,
      "loss": 2.1604,
      "step": 7
    },
    {
      "epoch": 0.00043723612116906006,
      "grad_norm": 1.1799358129501343,
      "learning_rate": 0.00016,
      "loss": 1.9058,
      "step": 8
    },
    {
      "epoch": 0.0004918906363151926,
      "grad_norm": 1.1987409591674805,
      "learning_rate": 0.00018,
      "loss": 2.057,
      "step": 9
    },
    {
      "epoch": 0.0004918906363151926,
      "eval_loss": 1.8708151578903198,
      "eval_runtime": 427.3376,
      "eval_samples_per_second": 18.028,
      "eval_steps_per_second": 9.014,
      "step": 9
    },
    {
      "epoch": 0.0005465451514613251,
      "grad_norm": 1.4058369398117065,
      "learning_rate": 0.0002,
      "loss": 1.7006,
      "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
}