lima/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 1000,
  "global_step": 155,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3225806451612903,
      "grad_norm": 0.40279635787010193,
      "learning_rate": 0.000125,
      "loss": 2.2735,
      "step": 10
    },
    {
      "epoch": 0.6451612903225806,
      "grad_norm": 0.31329983472824097,
      "learning_rate": 0.00019959162014075553,
      "loss": 2.1138,
      "step": 20
    },
    {
      "epoch": 0.967741935483871,
      "grad_norm": 0.2201918661594391,
      "learning_rate": 0.00019503556665478067,
      "loss": 2.0106,
      "step": 30
    },
    {
      "epoch": 1.2903225806451613,
      "grad_norm": 0.27235254645347595,
      "learning_rate": 0.00018564551148870563,
      "loss": 1.9603,
      "step": 40
    },
    {
      "epoch": 1.6129032258064515,
      "grad_norm": 0.2621012330055237,
      "learning_rate": 0.00017189908153577473,
      "loss": 2.0729,
      "step": 50
    },
    {
      "epoch": 1.935483870967742,
      "grad_norm": 0.2580893337726593,
      "learning_rate": 0.00015449549149872376,
      "loss": 1.9079,
      "step": 60
    },
    {
      "epoch": 2.258064516129032,
      "grad_norm": 0.371009886264801,
      "learning_rate": 0.00013431997820456592,
      "loss": 1.927,
      "step": 70
    },
    {
      "epoch": 2.5806451612903225,
      "grad_norm": 0.4627054035663605,
      "learning_rate": 0.00011239877286961122,
      "loss": 1.9111,
      "step": 80
    },
    {
      "epoch": 2.903225806451613,
      "grad_norm": 0.4472900331020355,
      "learning_rate": 8.984690165878921e-05,
      "loss": 1.8636,
      "step": 90
    },
    {
      "epoch": 3.225806451612903,
      "grad_norm": 0.602531373500824,
      "learning_rate": 6.781146967348284e-05,
      "loss": 1.8102,
      "step": 100
    },
    {
      "epoch": 3.5483870967741935,
      "grad_norm": 0.6399003863334656,
      "learning_rate": 4.7413313238324556e-05,
      "loss": 1.7998,
      "step": 110
    },
    {
      "epoch": 3.870967741935484,
      "grad_norm": 0.7003827691078186,
      "learning_rate": 2.968998835418174e-05,
      "loss": 1.7756,
      "step": 120
    },
    {
      "epoch": 4.193548387096774,
      "grad_norm": 0.5751145482063293,
      "learning_rate": 1.554299522021796e-05,
      "loss": 1.7402,
      "step": 130
    },
    {
      "epoch": 4.516129032258064,
      "grad_norm": 0.7806760668754578,
      "learning_rate": 5.691923259479093e-06,
      "loss": 1.7489,
      "step": 140
    },
    {
      "epoch": 4.838709677419355,
      "grad_norm": 0.7050411105155945,
      "learning_rate": 6.378490697611761e-07,
      "loss": 1.7256,
      "step": 150
    },
    {
      "epoch": 5.0,
      "step": 155,
      "total_flos": 7.90044384249774e+16,
      "train_loss": 1.902422603484123,
      "train_runtime": 278.0882,
      "train_samples_per_second": 17.387,
      "train_steps_per_second": 0.557
    }
  ],
  "logging_steps": 10,
  "max_steps": 155,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.90044384249774e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}