trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.985645933014354,
  "eval_steps": 500,
  "global_step": 78,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3827751196172249,
      "grad_norm": 3.3279682063032268,
      "learning_rate": 5e-06,
      "loss": 0.6611,
      "step": 10
    },
    {
      "epoch": 0.7655502392344498,
      "grad_norm": 0.9051411832795431,
      "learning_rate": 5e-06,
      "loss": 0.5553,
      "step": 20
    },
    {
      "epoch": 0.9952153110047847,
      "eval_loss": 0.49977537989616394,
      "eval_runtime": 27.5748,
      "eval_samples_per_second": 25.531,
      "eval_steps_per_second": 0.399,
      "step": 26
    },
    {
      "epoch": 1.1483253588516746,
      "grad_norm": 0.6319343232131018,
      "learning_rate": 5e-06,
      "loss": 0.558,
      "step": 30
    },
    {
      "epoch": 1.5311004784688995,
      "grad_norm": 0.600669196106336,
      "learning_rate": 5e-06,
      "loss": 0.4847,
      "step": 40
    },
    {
      "epoch": 1.9138755980861244,
      "grad_norm": 0.8550141582235415,
      "learning_rate": 5e-06,
      "loss": 0.4678,
      "step": 50
    },
    {
      "epoch": 1.9904306220095693,
      "eval_loss": 0.46811407804489136,
      "eval_runtime": 27.5558,
      "eval_samples_per_second": 25.548,
      "eval_steps_per_second": 0.399,
      "step": 52
    },
    {
      "epoch": 2.2966507177033493,
      "grad_norm": 0.7453595051934965,
      "learning_rate": 5e-06,
      "loss": 0.4895,
      "step": 60
    },
    {
      "epoch": 2.679425837320574,
      "grad_norm": 0.7300330189728759,
      "learning_rate": 5e-06,
      "loss": 0.4355,
      "step": 70
    },
    {
      "epoch": 2.985645933014354,
      "eval_loss": 0.45799314975738525,
      "eval_runtime": 27.079,
      "eval_samples_per_second": 25.998,
      "eval_steps_per_second": 0.406,
      "step": 78
    },
    {
      "epoch": 2.985645933014354,
      "step": 78,
      "total_flos": 130443525488640.0,
      "train_loss": 0.5125297094002749,
      "train_runtime": 4852.6983,
      "train_samples_per_second": 8.26,
      "train_steps_per_second": 0.016
    }
  ],
  "logging_steps": 10,
  "max_steps": 78,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 130443525488640.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}