checkpoints/checkpoint-32/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5565217391304348,
  "eval_steps": 4,
  "global_step": 32,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.017391304347826087,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 2.5696372985839844,
      "num_input_tokens_seen": 1494,
      "num_tokens": 1494.0,
      "step": 1,
      "train_runtime": 42.4619,
      "train_tokens_per_second": 35.185
    },
    {
      "epoch": 0.06956521739130435,
      "eval_loss": 5.31978178024292,
      "eval_num_tokens": 4980.0,
      "eval_runtime": 28.7144,
      "eval_samples_per_second": 1.045,
      "eval_steps_per_second": 0.279,
      "num_input_tokens_seen": 4980,
      "step": 4
    },
    {
      "epoch": 0.1391304347826087,
      "eval_loss": 5.31978178024292,
      "eval_num_tokens": 9004.0,
      "eval_runtime": 3.4006,
      "eval_samples_per_second": 8.822,
      "eval_steps_per_second": 2.353,
      "num_input_tokens_seen": 9004,
      "step": 8
    },
    {
      "epoch": 0.17391304347826086,
      "grad_norm": 28.1390438079834,
      "learning_rate": 0.0,
      "loss": 2.5477417839898004,
      "num_input_tokens_seen": 11145,
      "num_tokens": 11145.0,
      "step": 10,
      "train_runtime": 146.4448,
      "train_tokens_per_second": 76.104
    },
    {
      "epoch": 0.20869565217391303,
      "eval_loss": 1.7850792407989502,
      "eval_num_tokens": 13293.0,
      "eval_runtime": 3.4816,
      "eval_samples_per_second": 8.617,
      "eval_steps_per_second": 2.298,
      "num_input_tokens_seen": 13293,
      "step": 12
    },
    {
      "epoch": 0.2782608695652174,
      "eval_loss": 0.5965744256973267,
      "eval_num_tokens": 17742.0,
      "eval_runtime": 3.4391,
      "eval_samples_per_second": 8.723,
      "eval_steps_per_second": 2.326,
      "num_input_tokens_seen": 17742,
      "step": 16
    },
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 6.518080711364746,
      "learning_rate": 0.0001425726650926814,
      "loss": 0.7919160842895507,
      "num_input_tokens_seen": 20264,
      "num_tokens": 20264.0,
      "step": 20,
      "train_runtime": 241.9804,
      "train_tokens_per_second": 83.742
    },
    {
      "epoch": 0.34782608695652173,
      "eval_loss": 0.6307523250579834,
      "eval_num_tokens": 20264.0,
      "eval_runtime": 3.512,
      "eval_samples_per_second": 8.542,
      "eval_steps_per_second": 2.278,
      "num_input_tokens_seen": 20264,
      "step": 20
    },
    {
      "epoch": 0.41739130434782606,
      "eval_loss": 0.3135659396648407,
      "eval_num_tokens": 24990.0,
      "eval_runtime": 3.4341,
      "eval_samples_per_second": 8.736,
      "eval_steps_per_second": 2.33,
      "num_input_tokens_seen": 24990,
      "step": 24
    },
    {
      "epoch": 0.48695652173913045,
      "eval_loss": 0.4761301875114441,
      "eval_num_tokens": 28628.0,
      "eval_runtime": 3.4198,
      "eval_samples_per_second": 8.772,
      "eval_steps_per_second": 2.339,
      "num_input_tokens_seen": 28628,
      "step": 28
    },
    {
      "epoch": 0.5217391304347826,
      "grad_norm": 3.650630235671997,
      "learning_rate": 0.00011490240573865023,
      "loss": 0.22239422798156738,
      "num_input_tokens_seen": 30604,
      "num_tokens": 30604.0,
      "step": 30,
      "train_runtime": 326.8185,
      "train_tokens_per_second": 93.642
    },
    {
      "epoch": 0.5565217391304348,
      "eval_loss": 0.2599254846572876,
      "eval_num_tokens": 33017.0,
      "eval_runtime": 3.409,
      "eval_samples_per_second": 8.8,
      "eval_steps_per_second": 2.347,
      "num_input_tokens_seen": 33017,
      "step": 32
    }
  ],
  "logging_steps": 10,
  "max_steps": 58,
  "num_input_tokens_seen": 33017,
  "num_train_epochs": 1,
  "save_steps": 8,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 182806031368192.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}