trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.967741935483871,
  "eval_steps": 500,
  "global_step": 72,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.027649769585253458,
      "grad_norm": 0.12118180724620531,
      "learning_rate": 1.25e-06,
      "loss": 0.4052,
      "step": 1
    },
    {
      "epoch": 0.1382488479262673,
      "grad_norm": 0.09276164809571165,
      "learning_rate": 6.25e-06,
      "loss": 0.5046,
      "step": 5
    },
    {
      "epoch": 0.2764976958525346,
      "grad_norm": 0.12423129115518604,
      "learning_rate": 9.975923633360985e-06,
      "loss": 0.494,
      "step": 10
    },
    {
      "epoch": 0.4147465437788018,
      "grad_norm": 0.0915349647233371,
      "learning_rate": 9.707720325915105e-06,
      "loss": 0.4316,
      "step": 15
    },
    {
      "epoch": 0.5529953917050692,
      "grad_norm": 0.07446859498374453,
      "learning_rate": 9.157348061512728e-06,
      "loss": 0.432,
      "step": 20
    },
    {
      "epoch": 0.6912442396313364,
      "grad_norm": 0.06496916435703345,
      "learning_rate": 8.357794774235094e-06,
      "loss": 0.4402,
      "step": 25
    },
    {
      "epoch": 0.8294930875576036,
      "grad_norm": 0.06298369476648798,
      "learning_rate": 7.3569836841299905e-06,
      "loss": 0.4185,
      "step": 30
    },
    {
      "epoch": 0.967741935483871,
      "grad_norm": 0.06731429829180048,
      "learning_rate": 6.21490089951632e-06,
      "loss": 0.4048,
      "step": 35
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.3824561536312103,
      "eval_runtime": 19.7357,
      "eval_samples_per_second": 19.356,
      "eval_steps_per_second": 4.864,
      "step": 37
    },
    {
      "epoch": 1.0829493087557605,
      "grad_norm": 0.06374957595243645,
      "learning_rate": 5e-06,
      "loss": 0.3963,
      "step": 40
    },
    {
      "epoch": 1.2211981566820276,
      "grad_norm": 0.0680946365437224,
      "learning_rate": 3.7850991004836813e-06,
      "loss": 0.3484,
      "step": 45
    },
    {
      "epoch": 1.359447004608295,
      "grad_norm": 0.05004119075815043,
      "learning_rate": 2.6430163158700116e-06,
      "loss": 0.3416,
      "step": 50
    },
    {
      "epoch": 1.4976958525345623,
      "grad_norm": 0.05461700270757283,
      "learning_rate": 1.642205225764908e-06,
      "loss": 0.37,
      "step": 55
    },
    {
      "epoch": 1.6359447004608296,
      "grad_norm": 0.05050588248109686,
      "learning_rate": 8.426519384872733e-07,
      "loss": 0.3603,
      "step": 60
    },
    {
      "epoch": 1.7741935483870968,
      "grad_norm": 0.04335256475558046,
      "learning_rate": 2.9227967408489653e-07,
      "loss": 0.3524,
      "step": 65
    },
    {
      "epoch": 1.912442396313364,
      "grad_norm": 0.04869022142601223,
      "learning_rate": 2.4076366639015914e-08,
      "loss": 0.3643,
      "step": 70
    },
    {
      "epoch": 1.967741935483871,
      "eval_loss": 0.37427836656570435,
      "eval_runtime": 19.6704,
      "eval_samples_per_second": 19.42,
      "eval_steps_per_second": 4.88,
      "step": 72
    },
    {
      "epoch": 1.967741935483871,
      "step": 72,
      "total_flos": 1.7760167664602317e+17,
      "train_loss": 0.4003302885426415,
      "train_runtime": 975.808,
      "train_samples_per_second": 3.554,
      "train_steps_per_second": 0.074
    }
  ],
  "logging_steps": 5,
  "max_steps": 72,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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": 1.7760167664602317e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}