last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00071301247771836,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 7.130124777183601e-05,
      "grad_norm": 1.4019535779953003,
      "learning_rate": 2e-05,
      "loss": 3.2477,
      "step": 1
    },
    {
      "epoch": 7.130124777183601e-05,
      "eval_loss": 3.821178913116455,
      "eval_runtime": 354.566,
      "eval_samples_per_second": 16.657,
      "eval_steps_per_second": 8.328,
      "step": 1
    },
    {
      "epoch": 0.00014260249554367202,
      "grad_norm": 1.0410041809082031,
      "learning_rate": 4e-05,
      "loss": 3.0513,
      "step": 2
    },
    {
      "epoch": 0.000213903743315508,
      "grad_norm": 1.3710907697677612,
      "learning_rate": 6e-05,
      "loss": 3.5283,
      "step": 3
    },
    {
      "epoch": 0.000213903743315508,
      "eval_loss": 3.8168468475341797,
      "eval_runtime": 350.616,
      "eval_samples_per_second": 16.845,
      "eval_steps_per_second": 8.422,
      "step": 3
    },
    {
      "epoch": 0.00028520499108734403,
      "grad_norm": 1.2581191062927246,
      "learning_rate": 8e-05,
      "loss": 3.381,
      "step": 4
    },
    {
      "epoch": 0.00035650623885918,
      "grad_norm": 1.4203277826309204,
      "learning_rate": 0.0001,
      "loss": 3.5872,
      "step": 5
    },
    {
      "epoch": 0.000427807486631016,
      "grad_norm": 1.3418052196502686,
      "learning_rate": 0.00012,
      "loss": 3.9859,
      "step": 6
    },
    {
      "epoch": 0.000427807486631016,
      "eval_loss": 3.7826521396636963,
      "eval_runtime": 351.1152,
      "eval_samples_per_second": 16.821,
      "eval_steps_per_second": 8.41,
      "step": 6
    },
    {
      "epoch": 0.0004991087344028521,
      "grad_norm": 1.320258378982544,
      "learning_rate": 0.00014,
      "loss": 3.5131,
      "step": 7
    },
    {
      "epoch": 0.0005704099821746881,
      "grad_norm": 1.641695499420166,
      "learning_rate": 0.00016,
      "loss": 3.702,
      "step": 8
    },
    {
      "epoch": 0.0006417112299465241,
      "grad_norm": 1.4608796834945679,
      "learning_rate": 0.00018,
      "loss": 3.3921,
      "step": 9
    },
    {
      "epoch": 0.0006417112299465241,
      "eval_loss": 3.664571762084961,
      "eval_runtime": 349.8869,
      "eval_samples_per_second": 16.88,
      "eval_steps_per_second": 8.44,
      "step": 9
    },
    {
      "epoch": 0.00071301247771836,
      "grad_norm": 1.6834579706192017,
      "learning_rate": 0.0002,
      "loss": 4.0507,
      "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": 1628733135912960.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}