last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0221606648199446,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00221606648199446,
      "grad_norm": 2.1346476078033447,
      "learning_rate": 2e-05,
      "loss": 3.4392,
      "step": 1
    },
    {
      "epoch": 0.00221606648199446,
      "eval_loss": 0.9273197650909424,
      "eval_runtime": 10.2155,
      "eval_samples_per_second": 18.599,
      "eval_steps_per_second": 9.3,
      "step": 1
    },
    {
      "epoch": 0.00443213296398892,
      "grad_norm": 1.4991915225982666,
      "learning_rate": 4e-05,
      "loss": 2.8142,
      "step": 2
    },
    {
      "epoch": 0.006648199445983379,
      "grad_norm": 2.0667805671691895,
      "learning_rate": 6e-05,
      "loss": 4.1755,
      "step": 3
    },
    {
      "epoch": 0.006648199445983379,
      "eval_loss": 0.9239233136177063,
      "eval_runtime": 10.2054,
      "eval_samples_per_second": 18.618,
      "eval_steps_per_second": 9.309,
      "step": 3
    },
    {
      "epoch": 0.00886426592797784,
      "grad_norm": 1.5053385496139526,
      "learning_rate": 8e-05,
      "loss": 3.045,
      "step": 4
    },
    {
      "epoch": 0.0110803324099723,
      "grad_norm": 2.516897201538086,
      "learning_rate": 0.0001,
      "loss": 4.6529,
      "step": 5
    },
    {
      "epoch": 0.013296398891966758,
      "grad_norm": 3.6286537647247314,
      "learning_rate": 0.00012,
      "loss": 5.0229,
      "step": 6
    },
    {
      "epoch": 0.013296398891966758,
      "eval_loss": 0.8571804761886597,
      "eval_runtime": 10.2538,
      "eval_samples_per_second": 18.53,
      "eval_steps_per_second": 9.265,
      "step": 6
    },
    {
      "epoch": 0.01551246537396122,
      "grad_norm": 3.429811954498291,
      "learning_rate": 0.00014,
      "loss": 3.2235,
      "step": 7
    },
    {
      "epoch": 0.01772853185595568,
      "grad_norm": 3.1918835639953613,
      "learning_rate": 0.00016,
      "loss": 2.4192,
      "step": 8
    },
    {
      "epoch": 0.019944598337950138,
      "grad_norm": 3.344249725341797,
      "learning_rate": 0.00018,
      "loss": 1.9088,
      "step": 9
    },
    {
      "epoch": 0.019944598337950138,
      "eval_loss": 0.4230140745639801,
      "eval_runtime": 10.2409,
      "eval_samples_per_second": 18.553,
      "eval_steps_per_second": 9.277,
      "step": 9
    },
    {
      "epoch": 0.0221606648199446,
      "grad_norm": 5.451910495758057,
      "learning_rate": 0.0002,
      "loss": 2.2961,
      "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": 3166224305356800.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}