last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.028011204481792718,
  "eval_steps": 5,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0018674136321195146,
      "grad_norm": 11.52504825592041,
      "learning_rate": 2e-05,
      "loss": 8.6504,
      "step": 1
    },
    {
      "epoch": 0.0018674136321195146,
      "eval_loss": 8.316863059997559,
      "eval_runtime": 8.1714,
      "eval_samples_per_second": 27.658,
      "eval_steps_per_second": 13.829,
      "step": 1
    },
    {
      "epoch": 0.003734827264239029,
      "grad_norm": 9.959823608398438,
      "learning_rate": 4e-05,
      "loss": 7.6632,
      "step": 2
    },
    {
      "epoch": 0.0056022408963585435,
      "grad_norm": 7.954950332641602,
      "learning_rate": 6e-05,
      "loss": 8.8727,
      "step": 3
    },
    {
      "epoch": 0.007469654528478058,
      "grad_norm": 9.595186233520508,
      "learning_rate": 8e-05,
      "loss": 8.6758,
      "step": 4
    },
    {
      "epoch": 0.009337068160597572,
      "grad_norm": 10.132702827453613,
      "learning_rate": 0.0001,
      "loss": 8.1186,
      "step": 5
    },
    {
      "epoch": 0.009337068160597572,
      "eval_loss": 7.952444553375244,
      "eval_runtime": 7.5626,
      "eval_samples_per_second": 29.884,
      "eval_steps_per_second": 14.942,
      "step": 5
    },
    {
      "epoch": 0.011204481792717087,
      "grad_norm": 8.388951301574707,
      "learning_rate": 0.00012,
      "loss": 7.557,
      "step": 6
    },
    {
      "epoch": 0.013071895424836602,
      "grad_norm": 14.64582347869873,
      "learning_rate": 0.00014,
      "loss": 8.6447,
      "step": 7
    },
    {
      "epoch": 0.014939309056956116,
      "grad_norm": 12.940552711486816,
      "learning_rate": 0.00016,
      "loss": 6.7169,
      "step": 8
    },
    {
      "epoch": 0.01680672268907563,
      "grad_norm": 13.50047779083252,
      "learning_rate": 0.00018,
      "loss": 5.8296,
      "step": 9
    },
    {
      "epoch": 0.018674136321195144,
      "grad_norm": 16.245317459106445,
      "learning_rate": 0.0002,
      "loss": 6.6651,
      "step": 10
    },
    {
      "epoch": 0.018674136321195144,
      "eval_loss": 5.606720924377441,
      "eval_runtime": 7.6047,
      "eval_samples_per_second": 29.718,
      "eval_steps_per_second": 14.859,
      "step": 10
    },
    {
      "epoch": 0.02054154995331466,
      "grad_norm": 14.252686500549316,
      "learning_rate": 0.00019510565162951537,
      "loss": 5.4462,
      "step": 11
    },
    {
      "epoch": 0.022408963585434174,
      "grad_norm": 13.769002914428711,
      "learning_rate": 0.00018090169943749476,
      "loss": 5.9482,
      "step": 12
    },
    {
      "epoch": 0.02427637721755369,
      "grad_norm": 14.226924896240234,
      "learning_rate": 0.00015877852522924732,
      "loss": 5.444,
      "step": 13
    },
    {
      "epoch": 0.026143790849673203,
      "grad_norm": 12.833412170410156,
      "learning_rate": 0.00013090169943749476,
      "loss": 4.164,
      "step": 14
    },
    {
      "epoch": 0.028011204481792718,
      "grad_norm": 29.60838508605957,
      "learning_rate": 0.0001,
      "loss": 5.389,
      "step": 15
    },
    {
      "epoch": 0.028011204481792718,
      "eval_loss": 5.058675289154053,
      "eval_runtime": 7.6248,
      "eval_samples_per_second": 29.64,
      "eval_steps_per_second": 14.82,
      "step": 15
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1451583854346240.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}