trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1404,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.21367521367521367,
      "grad_norm": 0.3822082579135895,
      "learning_rate": 0.00019800000000000002,
      "loss": 1.858,
      "step": 100
    },
    {
      "epoch": 0.42735042735042733,
      "grad_norm": 0.309566468000412,
      "learning_rate": 0.00018481595092024542,
      "loss": 0.7733,
      "step": 200
    },
    {
      "epoch": 0.6410256410256411,
      "grad_norm": 0.3551405668258667,
      "learning_rate": 0.000169478527607362,
      "loss": 0.7457,
      "step": 300
    },
    {
      "epoch": 0.8547008547008547,
      "grad_norm": 0.3160478472709656,
      "learning_rate": 0.00015414110429447854,
      "loss": 0.7462,
      "step": 400
    },
    {
      "epoch": 1.0683760683760684,
      "grad_norm": 0.37799474596977234,
      "learning_rate": 0.00013880368098159512,
      "loss": 0.7241,
      "step": 500
    },
    {
      "epoch": 1.282051282051282,
      "grad_norm": 0.3522268235683441,
      "learning_rate": 0.00012346625766871167,
      "loss": 0.7086,
      "step": 600
    },
    {
      "epoch": 1.4957264957264957,
      "grad_norm": 0.35121089220046997,
      "learning_rate": 0.00010812883435582823,
      "loss": 0.7051,
      "step": 700
    },
    {
      "epoch": 1.7094017094017095,
      "grad_norm": 0.3671390414237976,
      "learning_rate": 9.279141104294478e-05,
      "loss": 0.7041,
      "step": 800
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 0.35119470953941345,
      "learning_rate": 7.745398773006135e-05,
      "loss": 0.7067,
      "step": 900
    },
    {
      "epoch": 2.1367521367521367,
      "grad_norm": 0.41583627462387085,
      "learning_rate": 6.211656441717791e-05,
      "loss": 0.6924,
      "step": 1000
    },
    {
      "epoch": 2.3504273504273505,
      "grad_norm": 0.4248015880584717,
      "learning_rate": 4.677914110429448e-05,
      "loss": 0.681,
      "step": 1100
    },
    {
      "epoch": 2.564102564102564,
      "grad_norm": 0.43060022592544556,
      "learning_rate": 3.1441717791411045e-05,
      "loss": 0.6762,
      "step": 1200
    },
    {
      "epoch": 2.7777777777777777,
      "grad_norm": 0.418743371963501,
      "learning_rate": 1.6104294478527606e-05,
      "loss": 0.6767,
      "step": 1300
    },
    {
      "epoch": 2.9914529914529915,
      "grad_norm": 0.4656442403793335,
      "learning_rate": 7.668711656441718e-07,
      "loss": 0.6761,
      "step": 1400
    }
  ],
  "logging_steps": 100,
  "max_steps": 1404,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.26438705281698e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}