last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004227882887644013,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00042278828876440123,
      "grad_norm": 10.807845115661621,
      "learning_rate": 2e-05,
      "loss": 4.4399,
      "step": 1
    },
    {
      "epoch": 0.00042278828876440123,
      "eval_loss": 1.0324617624282837,
      "eval_runtime": 56.199,
      "eval_samples_per_second": 17.723,
      "eval_steps_per_second": 8.861,
      "step": 1
    },
    {
      "epoch": 0.0008455765775288025,
      "grad_norm": 10.579981803894043,
      "learning_rate": 4e-05,
      "loss": 4.0622,
      "step": 2
    },
    {
      "epoch": 0.0012683648662932037,
      "grad_norm": 9.296797752380371,
      "learning_rate": 6e-05,
      "loss": 3.6015,
      "step": 3
    },
    {
      "epoch": 0.0012683648662932037,
      "eval_loss": 1.022390604019165,
      "eval_runtime": 49.7184,
      "eval_samples_per_second": 20.033,
      "eval_steps_per_second": 10.016,
      "step": 3
    },
    {
      "epoch": 0.001691153155057605,
      "grad_norm": 12.533465385437012,
      "learning_rate": 8e-05,
      "loss": 3.6269,
      "step": 4
    },
    {
      "epoch": 0.0021139414438220063,
      "grad_norm": 14.110212326049805,
      "learning_rate": 0.0001,
      "loss": 3.6889,
      "step": 5
    },
    {
      "epoch": 0.0025367297325864075,
      "grad_norm": 11.258315086364746,
      "learning_rate": 0.00012,
      "loss": 3.9806,
      "step": 6
    },
    {
      "epoch": 0.0025367297325864075,
      "eval_loss": 0.9941397905349731,
      "eval_runtime": 49.8123,
      "eval_samples_per_second": 19.995,
      "eval_steps_per_second": 9.998,
      "step": 6
    },
    {
      "epoch": 0.0029595180213508087,
      "grad_norm": 14.319456100463867,
      "learning_rate": 0.00014,
      "loss": 3.54,
      "step": 7
    },
    {
      "epoch": 0.00338230631011521,
      "grad_norm": 9.163406372070312,
      "learning_rate": 0.00016,
      "loss": 3.3994,
      "step": 8
    },
    {
      "epoch": 0.003805094598879611,
      "grad_norm": 11.519614219665527,
      "learning_rate": 0.00018,
      "loss": 4.0177,
      "step": 9
    },
    {
      "epoch": 0.003805094598879611,
      "eval_loss": 0.9554546475410461,
      "eval_runtime": 49.824,
      "eval_samples_per_second": 19.99,
      "eval_steps_per_second": 9.995,
      "step": 9
    },
    {
      "epoch": 0.004227882887644013,
      "grad_norm": 8.285616874694824,
      "learning_rate": 0.0002,
      "loss": 3.5485,
      "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": 3198626033565696.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}