checkpoint-4488/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 1000,
  "global_step": 4488,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0002228163992869875,
      "grad_norm": 5.737454891204834,
      "learning_rate": 0.0,
      "loss": 1.9528,
      "step": 1
    },
    {
      "epoch": 0.22281639928698752,
      "grad_norm": 3.56156587600708,
      "learning_rate": 8.903743315508023e-06,
      "loss": 1.6859,
      "step": 1000
    },
    {
      "epoch": 0.22281639928698752,
      "eval_cosine_accuracy": 0.9496266841888428,
      "eval_loss": 0.5503689050674438,
      "eval_runtime": 34.9667,
      "eval_samples_per_second": 271.944,
      "eval_steps_per_second": 1.087,
      "step": 1000
    },
    {
      "epoch": 0.44563279857397503,
      "grad_norm": 4.381109237670898,
      "learning_rate": 1.7816399286987524e-05,
      "loss": 1.1429,
      "step": 2000
    },
    {
      "epoch": 0.44563279857397503,
      "eval_cosine_accuracy": 0.9598275423049927,
      "eval_loss": 0.4875295162200928,
      "eval_runtime": 34.4488,
      "eval_samples_per_second": 276.033,
      "eval_steps_per_second": 1.103,
      "step": 2000
    },
    {
      "epoch": 0.6684491978609626,
      "grad_norm": 4.41341495513916,
      "learning_rate": 1.9252327193503666e-05,
      "loss": 0.7836,
      "step": 3000
    },
    {
      "epoch": 0.6684491978609626,
      "eval_cosine_accuracy": 0.9656115174293518,
      "eval_loss": 0.45152774453163147,
      "eval_runtime": 37.3805,
      "eval_samples_per_second": 254.384,
      "eval_steps_per_second": 1.017,
      "step": 3000
    },
    {
      "epoch": 0.8912655971479501,
      "grad_norm": 4.65971040725708,
      "learning_rate": 1.8262032085561498e-05,
      "loss": 0.601,
      "step": 4000
    },
    {
      "epoch": 0.8912655971479501,
      "eval_cosine_accuracy": 0.9693974256515503,
      "eval_loss": 0.4410804212093353,
      "eval_runtime": 36.7023,
      "eval_samples_per_second": 259.085,
      "eval_steps_per_second": 1.035,
      "step": 4000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 22440,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}