trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.9936,
  "eval_steps": 500,
  "global_step": 312,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.192,
      "grad_norm": 0.3010159432888031,
      "learning_rate": 3.191489361702128e-05,
      "loss": 2.2514,
      "step": 15
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.2521279752254486,
      "learning_rate": 6.382978723404256e-05,
      "loss": 2.1582,
      "step": 30
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.2557184100151062,
      "learning_rate": 9.574468085106384e-05,
      "loss": 1.8683,
      "step": 45
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.22737619280815125,
      "learning_rate": 0.00012765957446808513,
      "loss": 1.6607,
      "step": 60
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.19700799882411957,
      "learning_rate": 0.00015957446808510637,
      "loss": 1.4682,
      "step": 75
    },
    {
      "epoch": 1.152,
      "grad_norm": 0.2368774265050888,
      "learning_rate": 0.00019148936170212768,
      "loss": 1.3196,
      "step": 90
    },
    {
      "epoch": 1.3439999999999999,
      "grad_norm": 0.3574005365371704,
      "learning_rate": 0.00018990825688073394,
      "loss": 1.1283,
      "step": 105
    },
    {
      "epoch": 1.536,
      "grad_norm": 0.3291820287704468,
      "learning_rate": 0.0001761467889908257,
      "loss": 0.9629,
      "step": 120
    },
    {
      "epoch": 1.728,
      "grad_norm": 0.4235256612300873,
      "learning_rate": 0.00016238532110091745,
      "loss": 0.8761,
      "step": 135
    },
    {
      "epoch": 1.92,
      "grad_norm": 0.3522832691669464,
      "learning_rate": 0.00014862385321100919,
      "loss": 0.7764,
      "step": 150
    },
    {
      "epoch": 2.112,
      "grad_norm": 0.3914245367050171,
      "learning_rate": 0.00013486238532110092,
      "loss": 0.6717,
      "step": 165
    },
    {
      "epoch": 2.304,
      "grad_norm": 0.3502177894115448,
      "learning_rate": 0.00012110091743119268,
      "loss": 0.6329,
      "step": 180
    },
    {
      "epoch": 2.496,
      "grad_norm": 0.5064918398857117,
      "learning_rate": 0.0001073394495412844,
      "loss": 0.5551,
      "step": 195
    },
    {
      "epoch": 2.6879999999999997,
      "grad_norm": 0.4178985357284546,
      "learning_rate": 9.357798165137616e-05,
      "loss": 0.5132,
      "step": 210
    },
    {
      "epoch": 2.88,
      "grad_norm": 0.3729366958141327,
      "learning_rate": 7.98165137614679e-05,
      "loss": 0.4831,
      "step": 225
    },
    {
      "epoch": 3.072,
      "grad_norm": 0.30623266100883484,
      "learning_rate": 6.605504587155963e-05,
      "loss": 0.4484,
      "step": 240
    },
    {
      "epoch": 3.2640000000000002,
      "grad_norm": 0.31436726450920105,
      "learning_rate": 5.229357798165138e-05,
      "loss": 0.4255,
      "step": 255
    },
    {
      "epoch": 3.456,
      "grad_norm": 0.32084089517593384,
      "learning_rate": 3.8532110091743125e-05,
      "loss": 0.3731,
      "step": 270
    },
    {
      "epoch": 3.648,
      "grad_norm": 0.3452220857143402,
      "learning_rate": 2.4770642201834864e-05,
      "loss": 0.3786,
      "step": 285
    },
    {
      "epoch": 3.84,
      "grad_norm": 0.3483213782310486,
      "learning_rate": 1.1009174311926607e-05,
      "loss": 0.3988,
      "step": 300
    }
  ],
  "logging_steps": 15,
  "max_steps": 312,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "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": 4.447199037200794e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}