last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.005006466686136259,
  "eval_steps": 5,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00016688222287120864,
      "eval_loss": NaN,
      "eval_runtime": 383.9925,
      "eval_samples_per_second": 6.573,
      "eval_steps_per_second": 3.287,
      "step": 1
    },
    {
      "epoch": 0.0005006466686136259,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.0008344111143560433,
      "eval_loss": NaN,
      "eval_runtime": 384.8125,
      "eval_samples_per_second": 6.559,
      "eval_steps_per_second": 3.28,
      "step": 5
    },
    {
      "epoch": 0.0010012933372272519,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.0015019400058408777,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.0016688222287120865,
      "eval_loss": NaN,
      "eval_runtime": 384.6975,
      "eval_samples_per_second": 6.561,
      "eval_steps_per_second": 3.28,
      "step": 10
    },
    {
      "epoch": 0.0020025866744545037,
      "grad_norm": NaN,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.0025032333430681296,
      "grad_norm": NaN,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.0025032333430681296,
      "eval_loss": NaN,
      "eval_runtime": 384.7442,
      "eval_samples_per_second": 6.56,
      "eval_steps_per_second": 3.28,
      "step": 15
    },
    {
      "epoch": 0.0030038800116817554,
      "grad_norm": NaN,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.003337644457424173,
      "eval_loss": NaN,
      "eval_runtime": 384.672,
      "eval_samples_per_second": 6.561,
      "eval_steps_per_second": 3.281,
      "step": 20
    },
    {
      "epoch": 0.0035045266802953817,
      "grad_norm": NaN,
      "learning_rate": 8.435655349597689e-05,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.0040051733489090075,
      "grad_norm": NaN,
      "learning_rate": 4.12214747707527e-05,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.004172055571780216,
      "eval_loss": NaN,
      "eval_runtime": 384.9239,
      "eval_samples_per_second": 6.557,
      "eval_steps_per_second": 3.279,
      "step": 25
    },
    {
      "epoch": 0.004505820017522633,
      "grad_norm": NaN,
      "learning_rate": 1.0899347581163221e-05,
      "loss": 0.0,
      "step": 27
    },
    {
      "epoch": 0.005006466686136259,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.005006466686136259,
      "eval_loss": NaN,
      "eval_runtime": 384.5805,
      "eval_samples_per_second": 6.563,
      "eval_steps_per_second": 3.281,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.1006005802369024e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}