last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0012305420537746877,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00012305420537746877,
      "grad_norm": 0.8632257580757141,
      "learning_rate": 2e-05,
      "loss": 0.748,
      "step": 1
    },
    {
      "epoch": 0.00012305420537746877,
      "eval_loss": 0.879511833190918,
      "eval_runtime": 129.4136,
      "eval_samples_per_second": 26.442,
      "eval_steps_per_second": 13.221,
      "step": 1
    },
    {
      "epoch": 0.00024610841075493753,
      "grad_norm": 0.734817385673523,
      "learning_rate": 4e-05,
      "loss": 0.6559,
      "step": 2
    },
    {
      "epoch": 0.0003691626161324063,
      "grad_norm": 1.451820731163025,
      "learning_rate": 6e-05,
      "loss": 0.7293,
      "step": 3
    },
    {
      "epoch": 0.0003691626161324063,
      "eval_loss": 0.8606021404266357,
      "eval_runtime": 128.1351,
      "eval_samples_per_second": 26.706,
      "eval_steps_per_second": 13.353,
      "step": 3
    },
    {
      "epoch": 0.0004922168215098751,
      "grad_norm": 1.176847219467163,
      "learning_rate": 8e-05,
      "loss": 1.0256,
      "step": 4
    },
    {
      "epoch": 0.0006152710268873438,
      "grad_norm": 0.9159776568412781,
      "learning_rate": 0.0001,
      "loss": 0.7577,
      "step": 5
    },
    {
      "epoch": 0.0007383252322648126,
      "grad_norm": 0.7881878018379211,
      "learning_rate": 0.00012,
      "loss": 0.5812,
      "step": 6
    },
    {
      "epoch": 0.0007383252322648126,
      "eval_loss": 0.5984408259391785,
      "eval_runtime": 128.1993,
      "eval_samples_per_second": 26.693,
      "eval_steps_per_second": 13.346,
      "step": 6
    },
    {
      "epoch": 0.0008613794376422815,
      "grad_norm": 0.7859431505203247,
      "learning_rate": 0.00014,
      "loss": 0.4963,
      "step": 7
    },
    {
      "epoch": 0.0009844336430197501,
      "grad_norm": 1.107433795928955,
      "learning_rate": 0.00016,
      "loss": 0.44,
      "step": 8
    },
    {
      "epoch": 0.001107487848397219,
      "grad_norm": 1.0245170593261719,
      "learning_rate": 0.00018,
      "loss": 0.2713,
      "step": 9
    },
    {
      "epoch": 0.001107487848397219,
      "eval_loss": 0.1577470749616623,
      "eval_runtime": 128.0679,
      "eval_samples_per_second": 26.72,
      "eval_steps_per_second": 13.36,
      "step": 9
    },
    {
      "epoch": 0.0012305420537746877,
      "grad_norm": 1.437125563621521,
      "learning_rate": 0.0002,
      "loss": 0.1699,
      "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": 2358194416386048.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}