trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.098901098901099,
  "eval_steps": 30,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.10989010989010989,
      "grad_norm": 0.946593701839447,
      "learning_rate": 6.428571428571429e-05,
      "loss": 1.2769,
      "step": 10
    },
    {
      "epoch": 0.21978021978021978,
      "grad_norm": 0.5511431097984314,
      "learning_rate": 0.00013571428571428572,
      "loss": 0.3058,
      "step": 20
    },
    {
      "epoch": 0.32967032967032966,
      "grad_norm": 0.2574247121810913,
      "learning_rate": 0.00019999177886783194,
      "loss": 0.0791,
      "step": 30
    },
    {
      "epoch": 0.32967032967032966,
      "eval_loss": 0.047579921782016754,
      "eval_runtime": 19.2863,
      "eval_samples_per_second": 7.933,
      "eval_steps_per_second": 3.992,
      "step": 30
    },
    {
      "epoch": 0.43956043956043955,
      "grad_norm": 0.13821761310100555,
      "learning_rate": 0.0001990068775649202,
      "loss": 0.0461,
      "step": 40
    },
    {
      "epoch": 0.5494505494505495,
      "grad_norm": 0.1544637382030487,
      "learning_rate": 0.00019639628606958533,
      "loss": 0.0362,
      "step": 50
    },
    {
      "epoch": 0.6593406593406593,
      "grad_norm": 0.08936449140310287,
      "learning_rate": 0.00019220287022200707,
      "loss": 0.0329,
      "step": 60
    },
    {
      "epoch": 0.6593406593406593,
      "eval_loss": 0.03117775171995163,
      "eval_runtime": 18.6529,
      "eval_samples_per_second": 8.202,
      "eval_steps_per_second": 4.128,
      "step": 60
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.06332624703645706,
      "learning_rate": 0.00018649548579446936,
      "loss": 0.0313,
      "step": 70
    },
    {
      "epoch": 0.8791208791208791,
      "grad_norm": 0.08727087825536728,
      "learning_rate": 0.00017936784788148328,
      "loss": 0.0328,
      "step": 80
    },
    {
      "epoch": 0.989010989010989,
      "grad_norm": 0.07422881573438644,
      "learning_rate": 0.0001709369921005258,
      "loss": 0.0316,
      "step": 90
    },
    {
      "epoch": 0.989010989010989,
      "eval_loss": 0.028637176379561424,
      "eval_runtime": 18.6556,
      "eval_samples_per_second": 8.201,
      "eval_steps_per_second": 4.127,
      "step": 90
    },
    {
      "epoch": 1.098901098901099,
      "grad_norm": 0.04150444269180298,
      "learning_rate": 0.00016134135287043669,
      "loss": 0.0289,
      "step": 100
    }
  ],
  "logging_steps": 10,
  "max_steps": 273,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.458612072538829e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}