checkpoint-474/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 474,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2109704641350211,
      "grad_norm": 17.118331909179688,
      "learning_rate": 1.9156118143459917e-05,
      "loss": 4.0557,
      "step": 50
    },
    {
      "epoch": 0.4219409282700422,
      "grad_norm": 17.078466415405273,
      "learning_rate": 1.8312236286919833e-05,
      "loss": 4.018,
      "step": 100
    },
    {
      "epoch": 0.6329113924050633,
      "grad_norm": 18.140644073486328,
      "learning_rate": 1.746835443037975e-05,
      "loss": 3.9942,
      "step": 150
    },
    {
      "epoch": 0.8438818565400844,
      "grad_norm": 14.235407829284668,
      "learning_rate": 1.662447257383966e-05,
      "loss": 3.9127,
      "step": 200
    },
    {
      "epoch": 1.0,
      "eval_loss": 3.480090379714966,
      "eval_runtime": 253.6051,
      "eval_samples_per_second": 0.469,
      "eval_steps_per_second": 0.237,
      "step": 237
    },
    {
      "epoch": 1.0548523206751055,
      "grad_norm": 14.833487510681152,
      "learning_rate": 1.578059071729958e-05,
      "loss": 3.6803,
      "step": 250
    },
    {
      "epoch": 1.2658227848101267,
      "grad_norm": 18.978660583496094,
      "learning_rate": 1.4936708860759495e-05,
      "loss": 3.4012,
      "step": 300
    },
    {
      "epoch": 1.4767932489451476,
      "grad_norm": 18.297719955444336,
      "learning_rate": 1.4092827004219412e-05,
      "loss": 3.1245,
      "step": 350
    },
    {
      "epoch": 1.6877637130801688,
      "grad_norm": 16.072620391845703,
      "learning_rate": 1.3248945147679326e-05,
      "loss": 3.122,
      "step": 400
    },
    {
      "epoch": 1.8987341772151898,
      "grad_norm": 14.235777854919434,
      "learning_rate": 1.240506329113924e-05,
      "loss": 2.7533,
      "step": 450
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.393742084503174,
      "eval_runtime": 230.7613,
      "eval_samples_per_second": 0.516,
      "eval_steps_per_second": 0.26,
      "step": 474
    }
  ],
  "logging_steps": 50,
  "max_steps": 1185,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 249030441977856.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}