last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.003608805485384338,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00036088054853843375,
      "grad_norm": 4.020681858062744,
      "learning_rate": 2e-05,
      "loss": 8.5712,
      "step": 1
    },
    {
      "epoch": 0.00036088054853843375,
      "eval_loss": 2.053340196609497,
      "eval_runtime": 239.3472,
      "eval_samples_per_second": 4.876,
      "eval_steps_per_second": 2.44,
      "step": 1
    },
    {
      "epoch": 0.0007217610970768675,
      "grad_norm": 3.2305173873901367,
      "learning_rate": 4e-05,
      "loss": 7.4849,
      "step": 2
    },
    {
      "epoch": 0.0010826416456153013,
      "grad_norm": 3.2316172122955322,
      "learning_rate": 6e-05,
      "loss": 7.9669,
      "step": 3
    },
    {
      "epoch": 0.0010826416456153013,
      "eval_loss": 2.0378758907318115,
      "eval_runtime": 196.5802,
      "eval_samples_per_second": 5.937,
      "eval_steps_per_second": 2.971,
      "step": 3
    },
    {
      "epoch": 0.001443522194153735,
      "grad_norm": 3.5986013412475586,
      "learning_rate": 8e-05,
      "loss": 7.8219,
      "step": 4
    },
    {
      "epoch": 0.001804402742692169,
      "grad_norm": 3.2186686992645264,
      "learning_rate": 0.0001,
      "loss": 7.4097,
      "step": 5
    },
    {
      "epoch": 0.0021652832912306026,
      "grad_norm": 4.080005645751953,
      "learning_rate": 0.00012,
      "loss": 7.7574,
      "step": 6
    },
    {
      "epoch": 0.0021652832912306026,
      "eval_loss": 1.8460029363632202,
      "eval_runtime": 158.9944,
      "eval_samples_per_second": 7.34,
      "eval_steps_per_second": 3.673,
      "step": 6
    },
    {
      "epoch": 0.0025261638397690365,
      "grad_norm": 5.115031719207764,
      "learning_rate": 0.00014,
      "loss": 7.8949,
      "step": 7
    },
    {
      "epoch": 0.00288704438830747,
      "grad_norm": 5.0654191970825195,
      "learning_rate": 0.00016,
      "loss": 7.6231,
      "step": 8
    },
    {
      "epoch": 0.003247924936845904,
      "grad_norm": 3.7317137718200684,
      "learning_rate": 0.00018,
      "loss": 6.2799,
      "step": 9
    },
    {
      "epoch": 0.003247924936845904,
      "eval_loss": 1.464600920677185,
      "eval_runtime": 159.0893,
      "eval_samples_per_second": 7.336,
      "eval_steps_per_second": 3.671,
      "step": 9
    },
    {
      "epoch": 0.003608805485384338,
      "grad_norm": 4.158239364624023,
      "learning_rate": 0.0002,
      "loss": 5.7451,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "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": 5705610716774400.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}