last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.006126512482769184,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0006126512482769183,
      "grad_norm": 1.3412106037139893,
      "learning_rate": 2e-05,
      "loss": 1.2605,
      "step": 1
    },
    {
      "epoch": 0.0006126512482769183,
      "eval_loss": 1.2916122674942017,
      "eval_runtime": 31.9584,
      "eval_samples_per_second": 10.764,
      "eval_steps_per_second": 10.764,
      "step": 1
    },
    {
      "epoch": 0.0012253024965538366,
      "grad_norm": 0.9073652029037476,
      "learning_rate": 4e-05,
      "loss": 0.9145,
      "step": 2
    },
    {
      "epoch": 0.001837953744830755,
      "grad_norm": 1.8176641464233398,
      "learning_rate": 6e-05,
      "loss": 1.5518,
      "step": 3
    },
    {
      "epoch": 0.001837953744830755,
      "eval_loss": 1.2262399196624756,
      "eval_runtime": 31.294,
      "eval_samples_per_second": 10.993,
      "eval_steps_per_second": 10.993,
      "step": 3
    },
    {
      "epoch": 0.0024506049931076732,
      "grad_norm": 1.203653335571289,
      "learning_rate": 8e-05,
      "loss": 0.9995,
      "step": 4
    },
    {
      "epoch": 0.003063256241384592,
      "grad_norm": 0.8254466652870178,
      "learning_rate": 0.0001,
      "loss": 0.7087,
      "step": 5
    },
    {
      "epoch": 0.00367590748966151,
      "grad_norm": 1.2285630702972412,
      "learning_rate": 0.00012,
      "loss": 0.9726,
      "step": 6
    },
    {
      "epoch": 0.00367590748966151,
      "eval_loss": 1.025619387626648,
      "eval_runtime": 31.2768,
      "eval_samples_per_second": 10.999,
      "eval_steps_per_second": 10.999,
      "step": 6
    },
    {
      "epoch": 0.004288558737938429,
      "grad_norm": 1.058066964149475,
      "learning_rate": 0.00014,
      "loss": 0.8224,
      "step": 7
    },
    {
      "epoch": 0.0049012099862153465,
      "grad_norm": 1.0622285604476929,
      "learning_rate": 0.00016,
      "loss": 0.9098,
      "step": 8
    },
    {
      "epoch": 0.005513861234492265,
      "grad_norm": 1.496508240699768,
      "learning_rate": 0.00018,
      "loss": 0.8408,
      "step": 9
    },
    {
      "epoch": 0.005513861234492265,
      "eval_loss": 0.7476592659950256,
      "eval_runtime": 31.2832,
      "eval_samples_per_second": 10.996,
      "eval_steps_per_second": 10.996,
      "step": 9
    },
    {
      "epoch": 0.006126512482769184,
      "grad_norm": 1.366001009941101,
      "learning_rate": 0.0002,
      "loss": 0.8999,
      "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": 1306268836823040.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}