checkpoint-5516/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 1000,
  "global_step": 5516,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00036258158085569254,
      "grad_norm": 6.934128284454346,
      "learning_rate": 0.0,
      "loss": 3.917,
      "step": 1
    },
    {
      "epoch": 0.36258158085569253,
      "grad_norm": 6.089656352996826,
      "learning_rate": 1.2072507552870091e-05,
      "loss": 3.1519,
      "step": 1000
    },
    {
      "epoch": 0.36258158085569253,
      "eval_cosine_accuracy": 0.9487853646278381,
      "eval_loss": 0.5210903286933899,
      "eval_runtime": 35.26,
      "eval_samples_per_second": 269.682,
      "eval_steps_per_second": 1.078,
      "step": 1000
    },
    {
      "epoch": 0.7251631617113851,
      "grad_norm": 7.9518561363220215,
      "learning_rate": 1.896056806164073e-05,
      "loss": 1.9693,
      "step": 2000
    },
    {
      "epoch": 0.7251631617113851,
      "eval_cosine_accuracy": 0.9558312892913818,
      "eval_loss": 0.46612653136253357,
      "eval_runtime": 34.6788,
      "eval_samples_per_second": 274.202,
      "eval_steps_per_second": 1.096,
      "step": 2000
    },
    {
      "epoch": 1.0877447425670776,
      "grad_norm": 4.833303451538086,
      "learning_rate": 1.5938963589666113e-05,
      "loss": 1.6615,
      "step": 3000
    },
    {
      "epoch": 1.0877447425670776,
      "eval_cosine_accuracy": 0.9593017101287842,
      "eval_loss": 0.44890451431274414,
      "eval_runtime": 35.1553,
      "eval_samples_per_second": 270.485,
      "eval_steps_per_second": 1.081,
      "step": 3000
    },
    {
      "epoch": 1.4503263234227701,
      "grad_norm": 6.265031814575195,
      "learning_rate": 1.2917359117691495e-05,
      "loss": 1.7865,
      "step": 4000
    },
    {
      "epoch": 1.4503263234227701,
      "eval_cosine_accuracy": 0.9624566435813904,
      "eval_loss": 0.44137638807296753,
      "eval_runtime": 35.3681,
      "eval_samples_per_second": 268.858,
      "eval_steps_per_second": 1.074,
      "step": 4000
    },
    {
      "epoch": 1.8129079042784626,
      "grad_norm": 7.480532646179199,
      "learning_rate": 9.89877625018885e-06,
      "loss": 1.2745,
      "step": 5000
    },
    {
      "epoch": 1.8129079042784626,
      "eval_cosine_accuracy": 0.9640340805053711,
      "eval_loss": 0.4353458285331726,
      "eval_runtime": 34.313,
      "eval_samples_per_second": 277.125,
      "eval_steps_per_second": 1.107,
      "step": 5000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 8274,
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}