{
  "best_global_step": 41,
  "best_metric": 2.55751514,
  "best_model_checkpoint": "/home/ge65wed/PSLM/output/lora-8/v2-20250707-123813/checkpoint-41",
  "epoch": 1.0,
  "eval_steps": 25,
  "global_step": 41,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02450229709035222,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 3.8918616771698,
      "memory(GiB)": 4.25,
      "step": 1,
      "token_acc": 0.40435510887772197,
      "train_speed(iter/s)": 0.035339
    },
    {
      "epoch": 0.1225114854517611,
      "grad_norm": 1.6332721710205078,
      "learning_rate": 9.766981960274653e-05,
      "loss": 3.1261677742004395,
      "memory(GiB)": 7.41,
      "step": 5,
      "token_acc": 0.4092135437744503,
      "train_speed(iter/s)": 0.033213
    },
    {
      "epoch": 0.2450229709035222,
      "grad_norm": 1.4651521444320679,
      "learning_rate": 8.857445899109715e-05,
      "loss": 2.639118957519531,
      "memory(GiB)": 9.04,
      "step": 10,
      "token_acc": 0.4357703428335399,
      "train_speed(iter/s)": 0.0324
    },
    {
      "epoch": 0.3675344563552833,
      "grad_norm": 1.319446086883545,
      "learning_rate": 7.388599092561315e-05,
      "loss": 2.7252979278564453,
      "memory(GiB)": 10.78,
      "step": 15,
      "token_acc": 0.4232362500796635,
      "train_speed(iter/s)": 0.032144
    },
    {
      "epoch": 0.4900459418070444,
      "grad_norm": 1.1134439706802368,
      "learning_rate": 5.573417126992003e-05,
      "loss": 2.706107330322266,
      "memory(GiB)": 10.78,
      "step": 20,
      "token_acc": 0.42854686156491834,
      "train_speed(iter/s)": 0.031844
    },
    {
      "epoch": 0.6125574272588055,
      "grad_norm": 1.1963179111480713,
      "learning_rate": 3.675092489016693e-05,
      "loss": 2.6325620651245116,
      "memory(GiB)": 10.78,
      "step": 25,
      "token_acc": 0.435288414929714,
      "train_speed(iter/s)": 0.031685
    },
    {
      "epoch": 0.6125574272588055,
      "eval_loss": 2.5632457733154297,
      "eval_runtime": 3.8665,
      "eval_samples_per_second": 1.552,
      "eval_steps_per_second": 1.552,
      "step": 25
    },
    {
      "epoch": 0.7350689127105666,
      "grad_norm": 1.0305734872817993,
      "learning_rate": 1.9688729451668114e-05,
      "loss": 2.6473711013793944,
      "memory(GiB)": 10.78,
      "step": 30,
      "token_acc": 0.435996801377868,
      "train_speed(iter/s)": 0.031498
    },
    {
      "epoch": 0.8575803981623277,
      "grad_norm": 0.9539395570755005,
      "learning_rate": 7.0215196506399515e-06,
      "loss": 2.6621904373168945,
      "memory(GiB)": 10.78,
      "step": 35,
      "token_acc": 0.4297314996725606,
      "train_speed(iter/s)": 0.031512
    },
    {
      "epoch": 0.9800918836140888,
      "grad_norm": 0.994426965713501,
      "learning_rate": 5.859788109825793e-07,
      "loss": 2.680731010437012,
      "memory(GiB)": 10.78,
      "step": 40,
      "token_acc": 0.42303860523038606,
      "train_speed(iter/s)": 0.031482
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.5575151443481445,
      "eval_runtime": 3.8994,
      "eval_samples_per_second": 1.539,
      "eval_steps_per_second": 1.539,
      "step": 41
    }
  ],
  "logging_steps": 5,
  "max_steps": 41,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4668120944183040.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}