trainer_state.json

{
  "best_metric": 0.29447925,
  "best_model_checkpoint": "/home/patrickbarker/output/qwen2-vl-7b-instruct/v4-20241227-193713/checkpoint-35",
  "epoch": 4.4,
  "eval_steps": 200,
  "global_step": 35,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "acc": 0.78819823,
      "epoch": 0.13333333333333333,
      "grad_norm": 56.16291588852024,
      "learning_rate": 0.0,
      "loss": 1.50741768,
      "memory(GiB)": 54.69,
      "step": 1,
      "train_speed(iter/s)": 0.023135
    },
    {
      "acc": 0.8331359,
      "epoch": 0.6666666666666666,
      "grad_norm": 23.585686677807654,
      "learning_rate": 9.797485121585682e-06,
      "loss": 1.13953185,
      "memory(GiB)": 73.87,
      "step": 5,
      "train_speed(iter/s)": 0.030499
    },
    {
      "acc": 0.77300258,
      "epoch": 1.2666666666666666,
      "grad_norm": 9.184454728437117,
      "learning_rate": 8.930372447166567e-06,
      "loss": 0.82231588,
      "memory(GiB)": 73.87,
      "step": 10,
      "train_speed(iter/s)": 0.03335
    },
    {
      "acc": 0.89596729,
      "epoch": 1.9333333333333333,
      "grad_norm": 9.996269150167423,
      "learning_rate": 7.077367357502932e-06,
      "loss": 0.59573269,
      "memory(GiB)": 73.87,
      "step": 15,
      "train_speed(iter/s)": 0.033366
    },
    {
      "acc": 0.82209816,
      "epoch": 2.533333333333333,
      "grad_norm": 3.8888089608671805,
      "learning_rate": 4.762614211839201e-06,
      "loss": 0.45042038,
      "memory(GiB)": 73.87,
      "step": 20,
      "train_speed(iter/s)": 0.03415
    },
    {
      "acc": 0.82671137,
      "epoch": 3.1333333333333333,
      "grad_norm": 5.551126458017844,
      "learning_rate": 2.9236326424970685e-06,
      "loss": 0.38845139,
      "memory(GiB)": 73.87,
      "step": 25,
      "train_speed(iter/s)": 0.034641
    },
    {
      "acc": 0.92284222,
      "epoch": 3.8,
      "grad_norm": 3.1841046339599792,
      "learning_rate": 1.070627552833434e-06,
      "loss": 0.41313691,
      "memory(GiB)": 73.87,
      "step": 30,
      "train_speed(iter/s)": 0.034428
    },
    {
      "acc": 0.8380208,
      "epoch": 4.4,
      "grad_norm": 3.1924122632651946,
      "learning_rate": 9.134747803509838e-08,
      "loss": 0.31420252,
      "memory(GiB)": 73.87,
      "step": 35,
      "train_speed(iter/s)": 0.034743
    },
    {
      "epoch": 4.4,
      "eval_acc": 0.8957055214723927,
      "eval_loss": 0.29447925090789795,
      "eval_runtime": 5.161,
      "eval_samples_per_second": 1.163,
      "eval_steps_per_second": 0.194,
      "step": 35
    }
  ],
  "logging_steps": 5,
  "max_steps": 35,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 79586798469120.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}