checkpoint-3000/trainer_state.json

{
  "best_metric": 2.045022964477539,
  "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l55-l/checkpoint-2500",
  "epoch": 0.552689756816507,
  "eval_steps": 500,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09211495946941783,
      "grad_norm": 0.030529862269759178,
      "learning_rate": 4.8464750675509704e-05,
      "loss": 0.7076,
      "step": 500
    },
    {
      "epoch": 0.09211495946941783,
      "eval_loss": 2.07871675491333,
      "eval_runtime": 73.4415,
      "eval_samples_per_second": 16.435,
      "eval_steps_per_second": 2.056,
      "step": 500
    },
    {
      "epoch": 0.18422991893883567,
      "grad_norm": 0.02040918357670307,
      "learning_rate": 4.6929501351019406e-05,
      "loss": 0.7045,
      "step": 1000
    },
    {
      "epoch": 0.18422991893883567,
      "eval_loss": 2.0783257484436035,
      "eval_runtime": 74.6654,
      "eval_samples_per_second": 16.165,
      "eval_steps_per_second": 2.022,
      "step": 1000
    },
    {
      "epoch": 0.2763448784082535,
      "grad_norm": 0.8010361194610596,
      "learning_rate": 4.539425202652911e-05,
      "loss": 0.7106,
      "step": 1500
    },
    {
      "epoch": 0.2763448784082535,
      "eval_loss": 2.0473792552948,
      "eval_runtime": 75.4141,
      "eval_samples_per_second": 16.005,
      "eval_steps_per_second": 2.002,
      "step": 1500
    },
    {
      "epoch": 0.36845983787767134,
      "grad_norm": 1.1971429586410522,
      "learning_rate": 4.3859002702038817e-05,
      "loss": 0.704,
      "step": 2000
    },
    {
      "epoch": 0.36845983787767134,
      "eval_loss": 2.281594753265381,
      "eval_runtime": 76.3969,
      "eval_samples_per_second": 15.799,
      "eval_steps_per_second": 1.977,
      "step": 2000
    },
    {
      "epoch": 0.46057479734708917,
      "grad_norm": 1.5381091833114624,
      "learning_rate": 4.232375337754851e-05,
      "loss": 0.6996,
      "step": 2500
    },
    {
      "epoch": 0.46057479734708917,
      "eval_loss": 2.045022964477539,
      "eval_runtime": 76.3329,
      "eval_samples_per_second": 15.812,
      "eval_steps_per_second": 1.978,
      "step": 2500
    },
    {
      "epoch": 0.552689756816507,
      "grad_norm": 5.1257524490356445,
      "learning_rate": 4.078850405305822e-05,
      "loss": 0.6739,
      "step": 3000
    },
    {
      "epoch": 0.552689756816507,
      "eval_loss": 2.0969045162200928,
      "eval_runtime": 76.2547,
      "eval_samples_per_second": 15.829,
      "eval_steps_per_second": 1.98,
      "step": 3000
    }
  ],
  "logging_steps": 500,
  "max_steps": 16284,
  "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": 1079994570482520.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}