trainer_state.json

{
  "best_metric": 0.29381299,
  "best_model_checkpoint": "/home/patrickbarker/output/qwen2-vl-7b-instruct/v8-20241217-222833/checkpoint-45",
  "epoch": 5.0,
  "eval_steps": 200,
  "global_step": 45,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "acc": 0.7759673,
      "epoch": 0.1111111111111111,
      "grad_norm": 28.293013232594156,
      "learning_rate": 0.0,
      "loss": 1.42674351,
      "memory(GiB)": 54.56,
      "step": 1,
      "train_speed(iter/s)": 0.023402
    },
    {
      "acc": 0.78593183,
      "epoch": 0.5555555555555556,
      "grad_norm": 28.098530620260714,
      "learning_rate": 9.94415971571253e-06,
      "loss": 1.32285571,
      "memory(GiB)": 74.75,
      "step": 5,
      "train_speed(iter/s)": 0.030842
    },
    {
      "acc": 0.82505379,
      "epoch": 1.1111111111111112,
      "grad_norm": 33.872255803028175,
      "learning_rate": 9.330194006220303e-06,
      "loss": 1.00022011,
      "memory(GiB)": 74.75,
      "step": 10,
      "train_speed(iter/s)": 0.031979
    },
    {
      "acc": 0.85572233,
      "epoch": 1.6666666666666665,
      "grad_norm": 6.8998324139702065,
      "learning_rate": 8.11763726439274e-06,
      "loss": 0.75925498,
      "memory(GiB)": 74.75,
      "step": 15,
      "train_speed(iter/s)": 0.032466
    },
    {
      "acc": 0.88447227,
      "epoch": 2.2222222222222223,
      "grad_norm": 7.3832309753119185,
      "learning_rate": 6.4741284944673155e-06,
      "loss": 0.67884035,
      "memory(GiB)": 74.75,
      "step": 20,
      "train_speed(iter/s)": 0.032646
    },
    {
      "acc": 0.89219322,
      "epoch": 2.7777777777777777,
      "grad_norm": 5.543989291834036,
      "learning_rate": 4.626886897114674e-06,
      "loss": 0.58285484,
      "memory(GiB)": 74.75,
      "step": 25,
      "train_speed(iter/s)": 0.032807
    },
    {
      "acc": 0.92895489,
      "epoch": 3.3333333333333335,
      "grad_norm": 6.834702685745279,
      "learning_rate": 2.8312982462817686e-06,
      "loss": 0.39188585,
      "memory(GiB)": 74.75,
      "step": 30,
      "train_speed(iter/s)": 0.032872
    },
    {
      "acc": 0.91307697,
      "epoch": 3.888888888888889,
      "grad_norm": 4.746873592967143,
      "learning_rate": 1.3356071667867832e-06,
      "loss": 0.46585631,
      "memory(GiB)": 74.75,
      "step": 35,
      "train_speed(iter/s)": 0.032953
    },
    {
      "acc": 0.94487181,
      "epoch": 4.444444444444445,
      "grad_norm": 3.1089911576574254,
      "learning_rate": 3.465966936533004e-07,
      "loss": 0.32673092,
      "memory(GiB)": 74.75,
      "step": 40,
      "train_speed(iter/s)": 0.032983
    },
    {
      "acc": 0.94372597,
      "epoch": 5.0,
      "grad_norm": 4.679834617125414,
      "learning_rate": 1e-09,
      "loss": 0.31840396,
      "memory(GiB)": 74.75,
      "step": 45,
      "train_speed(iter/s)": 0.033034
    },
    {
      "epoch": 5.0,
      "eval_acc": 0.8853046594982079,
      "eval_loss": 0.29381299018859863,
      "eval_runtime": 23.5181,
      "eval_samples_per_second": 1.531,
      "eval_steps_per_second": 0.213,
      "step": 45
    }
  ],
  "logging_steps": 5,
  "max_steps": 45,
  "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": 108370957172736.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}