checkpoint-1107/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 1107,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09,
      "grad_norm": 0.728710412979126,
      "learning_rate": 0.00014935225390597272,
      "loss": 0.7697,
      "step": 100
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.4121594727039337,
      "learning_rate": 0.000143158760134902,
      "loss": 0.0249,
      "step": 200
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.27509087324142456,
      "learning_rate": 0.0001309204946874557,
      "loss": 0.0127,
      "step": 300
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.15149985253810883,
      "learning_rate": 0.00011372282825110337,
      "loss": 0.0085,
      "step": 400
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.4903504550457001,
      "learning_rate": 9.309096414187454e-05,
      "loss": 0.0052,
      "step": 500
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.01745792105793953,
      "learning_rate": 7.085467314127197e-05,
      "loss": 0.0042,
      "step": 600
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.1444382220506668,
      "learning_rate": 4.8986017266885955e-05,
      "loss": 0.0035,
      "step": 700
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.15122884511947632,
      "learning_rate": 2.9424454210849628e-05,
      "loss": 0.0031,
      "step": 800
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.13453546166419983,
      "learning_rate": 1.3904833346489733e-05,
      "loss": 0.0031,
      "step": 900
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.15443812310695648,
      "learning_rate": 3.8035377600492578e-06,
      "loss": 0.003,
      "step": 1000
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.011079126037657261,
      "learning_rate": 1.641745429760144e-08,
      "loss": 0.0028,
      "step": 1100
    }
  ],
  "logging_steps": 100,
  "max_steps": 1107,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 4.840885711798272e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}