last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.009363295880149813,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0009363295880149813,
      "grad_norm": 5.952914237976074,
      "learning_rate": 2e-05,
      "loss": 3.3495,
      "step": 1
    },
    {
      "epoch": 0.0009363295880149813,
      "eval_loss": 4.152853965759277,
      "eval_runtime": 13.5875,
      "eval_samples_per_second": 33.119,
      "eval_steps_per_second": 16.559,
      "step": 1
    },
    {
      "epoch": 0.0018726591760299626,
      "grad_norm": 6.230649471282959,
      "learning_rate": 4e-05,
      "loss": 4.1748,
      "step": 2
    },
    {
      "epoch": 0.0028089887640449437,
      "grad_norm": 5.637884616851807,
      "learning_rate": 6e-05,
      "loss": 3.4513,
      "step": 3
    },
    {
      "epoch": 0.0028089887640449437,
      "eval_loss": 4.100905895233154,
      "eval_runtime": 13.5708,
      "eval_samples_per_second": 33.159,
      "eval_steps_per_second": 16.58,
      "step": 3
    },
    {
      "epoch": 0.003745318352059925,
      "grad_norm": 5.062402725219727,
      "learning_rate": 8e-05,
      "loss": 3.49,
      "step": 4
    },
    {
      "epoch": 0.0046816479400749065,
      "grad_norm": 6.043950080871582,
      "learning_rate": 0.0001,
      "loss": 3.5806,
      "step": 5
    },
    {
      "epoch": 0.0056179775280898875,
      "grad_norm": 6.160160541534424,
      "learning_rate": 0.00012,
      "loss": 3.8251,
      "step": 6
    },
    {
      "epoch": 0.0056179775280898875,
      "eval_loss": 3.481220245361328,
      "eval_runtime": 13.7231,
      "eval_samples_per_second": 32.791,
      "eval_steps_per_second": 16.396,
      "step": 6
    },
    {
      "epoch": 0.006554307116104869,
      "grad_norm": 4.230264186859131,
      "learning_rate": 0.00014,
      "loss": 2.5014,
      "step": 7
    },
    {
      "epoch": 0.00749063670411985,
      "grad_norm": 6.001251220703125,
      "learning_rate": 0.00016,
      "loss": 3.6172,
      "step": 8
    },
    {
      "epoch": 0.008426966292134831,
      "grad_norm": 6.376860618591309,
      "learning_rate": 0.00018,
      "loss": 2.8885,
      "step": 9
    },
    {
      "epoch": 0.008426966292134831,
      "eval_loss": 2.7346646785736084,
      "eval_runtime": 13.7004,
      "eval_samples_per_second": 32.846,
      "eval_steps_per_second": 16.423,
      "step": 9
    },
    {
      "epoch": 0.009363295880149813,
      "grad_norm": 4.708912372589111,
      "learning_rate": 0.0002,
      "loss": 2.623,
      "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": 512599763976192.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}