checkpoint-2500/trainer_state.json

{
  "best_metric": 1.1902339458465576,
  "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e4l58-l/checkpoint-2500",
  "epoch": 0.46057479734708917,
  "eval_steps": 500,
  "global_step": 2500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09211495946941783,
      "grad_norm": 20.04043960571289,
      "learning_rate": 4.8848563006632274e-08,
      "loss": 0.3758,
      "step": 500
    },
    {
      "epoch": 0.09211495946941783,
      "eval_loss": 1.4184927940368652,
      "eval_runtime": 74.3562,
      "eval_samples_per_second": 16.233,
      "eval_steps_per_second": 2.031,
      "step": 500
    },
    {
      "epoch": 0.18422991893883567,
      "grad_norm": 74.25025939941406,
      "learning_rate": 4.7697126013264556e-08,
      "loss": 0.4103,
      "step": 1000
    },
    {
      "epoch": 0.18422991893883567,
      "eval_loss": 1.3500770330429077,
      "eval_runtime": 75.7013,
      "eval_samples_per_second": 15.944,
      "eval_steps_per_second": 1.995,
      "step": 1000
    },
    {
      "epoch": 0.2763448784082535,
      "grad_norm": 0.5102410912513733,
      "learning_rate": 4.6545689019896826e-08,
      "loss": 0.433,
      "step": 1500
    },
    {
      "epoch": 0.2763448784082535,
      "eval_loss": 1.2885024547576904,
      "eval_runtime": 75.197,
      "eval_samples_per_second": 16.051,
      "eval_steps_per_second": 2.008,
      "step": 1500
    },
    {
      "epoch": 0.36845983787767134,
      "grad_norm": 0.1667678952217102,
      "learning_rate": 4.539425202652911e-08,
      "loss": 0.3424,
      "step": 2000
    },
    {
      "epoch": 0.36845983787767134,
      "eval_loss": 1.239119052886963,
      "eval_runtime": 75.387,
      "eval_samples_per_second": 16.011,
      "eval_steps_per_second": 2.003,
      "step": 2000
    },
    {
      "epoch": 0.46057479734708917,
      "grad_norm": 343.8175048828125,
      "learning_rate": 4.4242815033161385e-08,
      "loss": 0.3645,
      "step": 2500
    },
    {
      "epoch": 0.46057479734708917,
      "eval_loss": 1.1902339458465576,
      "eval_runtime": 75.9899,
      "eval_samples_per_second": 15.884,
      "eval_steps_per_second": 1.987,
      "step": 2500
    }
  ],
  "logging_steps": 500,
  "max_steps": 21712,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "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": 900115394852520.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}