last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0016875856977112118,
  "eval_steps": 5,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 8.437928488556059e-05,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 8.437928488556059e-05,
      "eval_loss": NaN,
      "eval_runtime": 196.951,
      "eval_samples_per_second": 25.336,
      "eval_steps_per_second": 12.668,
      "step": 1
    },
    {
      "epoch": 0.00016875856977112118,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.0002531378546566818,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.00033751713954224236,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.00042189642442780296,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.00042189642442780296,
      "eval_loss": NaN,
      "eval_runtime": 195.6697,
      "eval_samples_per_second": 25.502,
      "eval_steps_per_second": 12.751,
      "step": 5
    },
    {
      "epoch": 0.0005062757093133636,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.0005906549941989242,
      "grad_norm": NaN,
      "learning_rate": 0.00014,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.0006750342790844847,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.0007594135639700454,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.0008437928488556059,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.0008437928488556059,
      "eval_loss": NaN,
      "eval_runtime": 196.4371,
      "eval_samples_per_second": 25.403,
      "eval_steps_per_second": 12.701,
      "step": 10
    },
    {
      "epoch": 0.0009281721337411666,
      "grad_norm": NaN,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.0010125514186267272,
      "grad_norm": NaN,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.0010969307035122878,
      "grad_norm": NaN,
      "learning_rate": 0.00015877852522924732,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.0011813099883978483,
      "grad_norm": NaN,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.0012656892732834089,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.0012656892732834089,
      "eval_loss": NaN,
      "eval_runtime": 196.0993,
      "eval_samples_per_second": 25.446,
      "eval_steps_per_second": 12.723,
      "step": 15
    },
    {
      "epoch": 0.0013500685581689694,
      "grad_norm": NaN,
      "learning_rate": 6.909830056250527e-05,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 0.0014344478430545302,
      "grad_norm": NaN,
      "learning_rate": 4.12214747707527e-05,
      "loss": 0.0,
      "step": 17
    },
    {
      "epoch": 0.0015188271279400908,
      "grad_norm": NaN,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.0016032064128256513,
      "grad_norm": NaN,
      "learning_rate": 4.8943483704846475e-06,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 0.0016875856977112118,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.0016875856977112118,
      "eval_loss": NaN,
      "eval_runtime": 195.9674,
      "eval_samples_per_second": 25.463,
      "eval_steps_per_second": 12.732,
      "step": 20
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2057373417996288.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}