last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.013908205841446454,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0013908205841446453,
      "grad_norm": 0.8333961367607117,
      "learning_rate": 2e-05,
      "loss": 1.3831,
      "step": 1
    },
    {
      "epoch": 0.0013908205841446453,
      "eval_loss": 1.014951229095459,
      "eval_runtime": 4.9385,
      "eval_samples_per_second": 30.779,
      "eval_steps_per_second": 30.779,
      "step": 1
    },
    {
      "epoch": 0.0027816411682892906,
      "grad_norm": 0.9491351246833801,
      "learning_rate": 4e-05,
      "loss": 1.1721,
      "step": 2
    },
    {
      "epoch": 0.004172461752433936,
      "grad_norm": 0.7000308036804199,
      "learning_rate": 6e-05,
      "loss": 0.8161,
      "step": 3
    },
    {
      "epoch": 0.004172461752433936,
      "eval_loss": 1.0096120834350586,
      "eval_runtime": 4.9607,
      "eval_samples_per_second": 30.641,
      "eval_steps_per_second": 30.641,
      "step": 3
    },
    {
      "epoch": 0.005563282336578581,
      "grad_norm": 1.0017286539077759,
      "learning_rate": 8e-05,
      "loss": 0.8552,
      "step": 4
    },
    {
      "epoch": 0.006954102920723227,
      "grad_norm": 1.0567864179611206,
      "learning_rate": 0.0001,
      "loss": 0.9583,
      "step": 5
    },
    {
      "epoch": 0.008344923504867872,
      "grad_norm": 0.7813672423362732,
      "learning_rate": 0.00012,
      "loss": 0.8737,
      "step": 6
    },
    {
      "epoch": 0.008344923504867872,
      "eval_loss": 0.9795357584953308,
      "eval_runtime": 5.1418,
      "eval_samples_per_second": 29.562,
      "eval_steps_per_second": 29.562,
      "step": 6
    },
    {
      "epoch": 0.009735744089012517,
      "grad_norm": 0.6960843801498413,
      "learning_rate": 0.00014,
      "loss": 0.7209,
      "step": 7
    },
    {
      "epoch": 0.011126564673157162,
      "grad_norm": 0.6831668615341187,
      "learning_rate": 0.00016,
      "loss": 0.6779,
      "step": 8
    },
    {
      "epoch": 0.012517385257301807,
      "grad_norm": 0.4121267795562744,
      "learning_rate": 0.00018,
      "loss": 0.5585,
      "step": 9
    },
    {
      "epoch": 0.012517385257301807,
      "eval_loss": 0.9398213624954224,
      "eval_runtime": 5.0845,
      "eval_samples_per_second": 29.895,
      "eval_steps_per_second": 29.895,
      "step": 9
    },
    {
      "epoch": 0.013908205841446454,
      "grad_norm": 0.8026168346405029,
      "learning_rate": 0.0002,
      "loss": 0.8507,
      "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": 258586429095936.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}