trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9491525423728815,
  "eval_steps": 500,
  "global_step": 87,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3389830508474576,
      "grad_norm": 3.373545556197121,
      "learning_rate": 5e-06,
      "loss": 1.2216,
      "step": 10
    },
    {
      "epoch": 0.6779661016949152,
      "grad_norm": 1.2990332547586303,
      "learning_rate": 5e-06,
      "loss": 1.1268,
      "step": 20
    },
    {
      "epoch": 0.9830508474576272,
      "eval_loss": 1.0789530277252197,
      "eval_runtime": 32.501,
      "eval_samples_per_second": 24.399,
      "eval_steps_per_second": 0.4,
      "step": 29
    },
    {
      "epoch": 1.0169491525423728,
      "grad_norm": 1.5477571699054125,
      "learning_rate": 5e-06,
      "loss": 1.1342,
      "step": 30
    },
    {
      "epoch": 1.3559322033898304,
      "grad_norm": 1.0602443282760596,
      "learning_rate": 5e-06,
      "loss": 1.0292,
      "step": 40
    },
    {
      "epoch": 1.694915254237288,
      "grad_norm": 1.5678600443481705,
      "learning_rate": 5e-06,
      "loss": 1.015,
      "step": 50
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.044095754623413,
      "eval_runtime": 32.312,
      "eval_samples_per_second": 24.542,
      "eval_steps_per_second": 0.402,
      "step": 59
    },
    {
      "epoch": 2.0338983050847457,
      "grad_norm": 1.5189371025726972,
      "learning_rate": 5e-06,
      "loss": 1.0548,
      "step": 60
    },
    {
      "epoch": 2.3728813559322033,
      "grad_norm": 0.9125182340525109,
      "learning_rate": 5e-06,
      "loss": 0.9598,
      "step": 70
    },
    {
      "epoch": 2.711864406779661,
      "grad_norm": 0.8112693026166728,
      "learning_rate": 5e-06,
      "loss": 0.9549,
      "step": 80
    },
    {
      "epoch": 2.9491525423728815,
      "eval_loss": 1.0310115814208984,
      "eval_runtime": 31.8954,
      "eval_samples_per_second": 24.863,
      "eval_steps_per_second": 0.408,
      "step": 87
    },
    {
      "epoch": 2.9491525423728815,
      "step": 87,
      "total_flos": 145518860697600.0,
      "train_loss": 1.0528919669403427,
      "train_runtime": 5360.6063,
      "train_samples_per_second": 8.421,
      "train_steps_per_second": 0.016
    }
  ],
  "logging_steps": 10,
  "max_steps": 87,
  "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": 145518860697600.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}