checkpoint-7989/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 1000,
  "global_step": 7989,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0003755163349605708,
      "grad_norm": 6.263205528259277,
      "learning_rate": 0.0,
      "loss": 3.7128,
      "step": 1
    },
    {
      "epoch": 0.3755163349605708,
      "grad_norm": 3.8261115550994873,
      "learning_rate": 7.502816372512205e-06,
      "loss": 2.8207,
      "step": 1000
    },
    {
      "epoch": 0.3755163349605708,
      "eval_cosine_accuracy": 0.9523609280586243,
      "eval_loss": 1.359626054763794,
      "eval_runtime": 22.5746,
      "eval_samples_per_second": 421.226,
      "eval_steps_per_second": 1.683,
      "step": 1000
    },
    {
      "epoch": 0.7510326699211416,
      "grad_norm": 3.8652374744415283,
      "learning_rate": 1.5013143071723622e-05,
      "loss": 2.2515,
      "step": 2000
    },
    {
      "epoch": 0.7510326699211416,
      "eval_cosine_accuracy": 0.9578294157981873,
      "eval_loss": 1.3198553323745728,
      "eval_runtime": 22.5254,
      "eval_samples_per_second": 422.145,
      "eval_steps_per_second": 1.687,
      "step": 2000
    },
    {
      "epoch": 1.1265490048817124,
      "grad_norm": 3.8326590061187744,
      "learning_rate": 1.9372887720615848e-05,
      "loss": 1.8906,
      "step": 3000
    },
    {
      "epoch": 1.1265490048817124,
      "eval_cosine_accuracy": 0.9598275423049927,
      "eval_loss": 1.3083600997924805,
      "eval_runtime": 22.9203,
      "eval_samples_per_second": 414.873,
      "eval_steps_per_second": 1.658,
      "step": 3000
    },
    {
      "epoch": 1.5020653398422832,
      "grad_norm": 3.9242522716522217,
      "learning_rate": 1.7495306045812995e-05,
      "loss": 1.7993,
      "step": 4000
    },
    {
      "epoch": 1.5020653398422832,
      "eval_cosine_accuracy": 0.9630876183509827,
      "eval_loss": 1.272866129875183,
      "eval_runtime": 22.2844,
      "eval_samples_per_second": 426.712,
      "eval_steps_per_second": 1.705,
      "step": 4000
    },
    {
      "epoch": 1.8775816748028538,
      "grad_norm": 3.9837265014648438,
      "learning_rate": 1.561772437101014e-05,
      "loss": 1.6857,
      "step": 5000
    },
    {
      "epoch": 1.8775816748028538,
      "eval_cosine_accuracy": 0.9651908874511719,
      "eval_loss": 1.2809278964996338,
      "eval_runtime": 22.5564,
      "eval_samples_per_second": 421.565,
      "eval_steps_per_second": 1.685,
      "step": 5000
    },
    {
      "epoch": 2.253098009763425,
      "grad_norm": 3.77929425239563,
      "learning_rate": 1.3742020277882089e-05,
      "loss": 1.5251,
      "step": 6000
    },
    {
      "epoch": 2.253098009763425,
      "eval_cosine_accuracy": 0.9658218622207642,
      "eval_loss": 1.2626943588256836,
      "eval_runtime": 23.9918,
      "eval_samples_per_second": 396.343,
      "eval_steps_per_second": 1.584,
      "step": 6000
    },
    {
      "epoch": 2.6286143447239954,
      "grad_norm": 3.7737784385681152,
      "learning_rate": 1.1864438603079236e-05,
      "loss": 1.5447,
      "step": 7000
    },
    {
      "epoch": 2.6286143447239954,
      "eval_cosine_accuracy": 0.965927004814148,
      "eval_loss": 1.2667568922042847,
      "eval_runtime": 22.5788,
      "eval_samples_per_second": 421.147,
      "eval_steps_per_second": 1.683,
      "step": 7000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 13315,
  "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
}