checkpoint-300/trainer_state.json

{
  "best_global_step": 300,
  "best_metric": 0.33731985092163086,
  "best_model_checkpoint": "trained_models/full_ft/smolvlm256m_fullft_qwen2_5_vl_3b_gt_20250503/checkpoint-300",
  "epoch": 3.8051118210862622,
  "eval_steps": 100,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.268370607028754,
      "grad_norm": 5.65625,
      "learning_rate": 9.315068493150684e-05,
      "loss": 0.5957,
      "step": 100
    },
    {
      "epoch": 1.268370607028754,
      "eval_loss": 0.42437922954559326,
      "eval_runtime": 38.726,
      "eval_samples_per_second": 8.96,
      "eval_steps_per_second": 4.493,
      "step": 100
    },
    {
      "epoch": 2.536741214057508,
      "grad_norm": 3.4375,
      "learning_rate": 7.945205479452055e-05,
      "loss": 0.3141,
      "step": 200
    },
    {
      "epoch": 2.536741214057508,
      "eval_loss": 0.3681740462779999,
      "eval_runtime": 38.3364,
      "eval_samples_per_second": 9.051,
      "eval_steps_per_second": 4.539,
      "step": 200
    },
    {
      "epoch": 3.8051118210862622,
      "grad_norm": 2.59375,
      "learning_rate": 6.575342465753424e-05,
      "loss": 0.231,
      "step": 300
    },
    {
      "epoch": 3.8051118210862622,
      "eval_loss": 0.33731985092163086,
      "eval_runtime": 38.4861,
      "eval_samples_per_second": 9.016,
      "eval_steps_per_second": 4.521,
      "step": 300
    }
  ],
  "logging_steps": 100,
  "max_steps": 780,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4121060856052608.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}