checkpoint-5576/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9992831541218639,
  "eval_steps": 1000,
  "global_step": 5576,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0003586800573888092,
      "grad_norm": 5.261780738830566,
      "learning_rate": 0.0,
      "loss": 1.248,
      "step": 1
    },
    {
      "epoch": 0.3586800573888092,
      "grad_norm": 6.529983043670654,
      "learning_rate": 4.562574730968513e-05,
      "loss": 1.223,
      "step": 1000
    },
    {
      "epoch": 0.3586800573888092,
      "eval_cosine_accuracy": 0.9625617861747742,
      "eval_loss": 0.4690546989440918,
      "eval_runtime": 59.1908,
      "eval_samples_per_second": 160.65,
      "eval_steps_per_second": 0.642,
      "step": 1000
    },
    {
      "epoch": 0.7173601147776184,
      "grad_norm": 5.523382186889648,
      "learning_rate": 3.5661618174571546e-05,
      "loss": 0.8193,
      "step": 2000
    },
    {
      "epoch": 0.7173601147776184,
      "eval_cosine_accuracy": 0.9603533744812012,
      "eval_loss": 0.44776350259780884,
      "eval_runtime": 56.9539,
      "eval_samples_per_second": 166.959,
      "eval_steps_per_second": 0.667,
      "step": 2000
    },
    {
      "epoch": 1.0759856630824374,
      "grad_norm": 4.964259147644043,
      "learning_rate": 2.569748903945795e-05,
      "loss": 0.6847,
      "step": 3000
    },
    {
      "epoch": 1.0759856630824374,
      "eval_cosine_accuracy": 0.9650856852531433,
      "eval_loss": 0.44899600744247437,
      "eval_runtime": 57.5082,
      "eval_samples_per_second": 165.35,
      "eval_steps_per_second": 0.661,
      "step": 3000
    },
    {
      "epoch": 1.4344086021505376,
      "grad_norm": 4.930419445037842,
      "learning_rate": 1.5743324033479476e-05,
      "loss": 0.7661,
      "step": 4000
    },
    {
      "epoch": 1.4344086021505376,
      "eval_cosine_accuracy": 0.9645599126815796,
      "eval_loss": 0.4399823844432831,
      "eval_runtime": 57.6844,
      "eval_samples_per_second": 164.845,
      "eval_steps_per_second": 0.659,
      "step": 4000
    },
    {
      "epoch": 1.792831541218638,
      "grad_norm": 5.024491310119629,
      "learning_rate": 5.779194898365883e-06,
      "loss": 0.7319,
      "step": 5000
    },
    {
      "epoch": 1.792831541218638,
      "eval_cosine_accuracy": 0.9647701978683472,
      "eval_loss": 0.4436001181602478,
      "eval_runtime": 59.2822,
      "eval_samples_per_second": 160.402,
      "eval_steps_per_second": 0.641,
      "step": 5000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 5576,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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": 0.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}