last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.02386634844868735,
  "eval_steps": 1,
  "global_step": 5,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00477326968973747,
      "grad_norm": 0.7191289663314819,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.9479,
      "step": 1
    },
    {
      "epoch": 0.00477326968973747,
      "eval_loss": 0.9769248962402344,
      "eval_runtime": 24.6512,
      "eval_samples_per_second": 7.18,
      "eval_steps_per_second": 1.825,
      "step": 1
    },
    {
      "epoch": 0.00954653937947494,
      "grad_norm": 0.507362425327301,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.825,
      "step": 2
    },
    {
      "epoch": 0.00954653937947494,
      "eval_loss": 0.9766976833343506,
      "eval_runtime": 24.7531,
      "eval_samples_per_second": 7.151,
      "eval_steps_per_second": 1.818,
      "step": 2
    },
    {
      "epoch": 0.014319809069212411,
      "grad_norm": 0.7874433398246765,
      "learning_rate": 6e-06,
      "loss": 0.8523,
      "step": 3
    },
    {
      "epoch": 0.014319809069212411,
      "eval_loss": 0.9762276411056519,
      "eval_runtime": 24.8247,
      "eval_samples_per_second": 7.13,
      "eval_steps_per_second": 1.813,
      "step": 3
    },
    {
      "epoch": 0.01909307875894988,
      "grad_norm": 0.9041501879692078,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.0266,
      "step": 4
    },
    {
      "epoch": 0.01909307875894988,
      "eval_loss": 0.9775993824005127,
      "eval_runtime": 24.8288,
      "eval_samples_per_second": 7.129,
      "eval_steps_per_second": 1.812,
      "step": 4
    },
    {
      "epoch": 0.02386634844868735,
      "grad_norm": 0.7414589524269104,
      "learning_rate": 1e-05,
      "loss": 0.8384,
      "step": 5
    },
    {
      "epoch": 0.02386634844868735,
      "eval_loss": 0.9767054915428162,
      "eval_runtime": 24.8805,
      "eval_samples_per_second": 7.114,
      "eval_steps_per_second": 1.809,
      "step": 5
    }
  ],
  "logging_steps": 1,
  "max_steps": 5,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 20,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1935445139128320.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}