trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9826435246995995,
  "eval_steps": 3,
  "global_step": 46,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06408544726301736,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 1.0852,
      "step": 3
    },
    {
      "epoch": 0.12817089452603472,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 1.0936,
      "step": 6
    },
    {
      "epoch": 0.19225634178905207,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 1.0951,
      "step": 9
    },
    {
      "epoch": 0.25634178905206945,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 1.0732,
      "step": 12
    },
    {
      "epoch": 0.3204272363150868,
      "grad_norm": 0.253641277551651,
      "learning_rate": 0.001414213562373095,
      "loss": 1.0803,
      "step": 15
    },
    {
      "epoch": 0.38451268357810414,
      "grad_norm": 0.3905181288719177,
      "learning_rate": 0.0011547005383792518,
      "loss": 1.2341,
      "step": 18
    },
    {
      "epoch": 0.4485981308411215,
      "grad_norm": 0.34546247124671936,
      "learning_rate": 0.0008164965809277262,
      "loss": 1.0057,
      "step": 21
    },
    {
      "epoch": 0.5126835781041389,
      "grad_norm": 0.2852560877799988,
      "learning_rate": 0.0006666666666666666,
      "loss": 0.7892,
      "step": 24
    },
    {
      "epoch": 0.5767690253671562,
      "grad_norm": 0.23831070959568024,
      "learning_rate": 0.0005773502691896259,
      "loss": 0.769,
      "step": 27
    },
    {
      "epoch": 0.6408544726301736,
      "grad_norm": 0.24124804139137268,
      "learning_rate": 0.0005163977794943222,
      "loss": 0.7525,
      "step": 30
    },
    {
      "epoch": 0.7049399198931909,
      "grad_norm": 0.23033007979393005,
      "learning_rate": 0.00047140452079103175,
      "loss": 0.7335,
      "step": 33
    },
    {
      "epoch": 0.7690253671562083,
      "grad_norm": 0.20133820176124573,
      "learning_rate": 0.0004364357804719848,
      "loss": 0.7292,
      "step": 36
    },
    {
      "epoch": 0.8331108144192256,
      "grad_norm": 0.22399117052555084,
      "learning_rate": 0.0004082482904638631,
      "loss": 0.7211,
      "step": 39
    },
    {
      "epoch": 0.897196261682243,
      "grad_norm": 0.2745206654071808,
      "learning_rate": 0.00038490017945975053,
      "loss": 0.6917,
      "step": 42
    },
    {
      "epoch": 0.9612817089452603,
      "grad_norm": 0.20307739078998566,
      "learning_rate": 0.00036514837167011074,
      "loss": 0.7122,
      "step": 45
    },
    {
      "epoch": 0.9826435246995995,
      "step": 46,
      "total_flos": 4.671973340489974e+17,
      "train_loss": 0.8987806633762692,
      "train_runtime": 500.6131,
      "train_samples_per_second": 11.961,
      "train_steps_per_second": 0.092
    }
  ],
  "logging_steps": 3,
  "max_steps": 46,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.671973340489974e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}