last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.005824111822947,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005824111822947001,
      "grad_norm": 4.431249618530273,
      "learning_rate": 2e-05,
      "loss": 2.441,
      "step": 1
    },
    {
      "epoch": 0.0005824111822947001,
      "eval_loss": 2.366206645965576,
      "eval_runtime": 109.5177,
      "eval_samples_per_second": 6.602,
      "eval_steps_per_second": 3.305,
      "step": 1
    },
    {
      "epoch": 0.0011648223645894002,
      "grad_norm": 3.182062864303589,
      "learning_rate": 4e-05,
      "loss": 2.416,
      "step": 2
    },
    {
      "epoch": 0.0017472335468841002,
      "grad_norm": 3.0722639560699463,
      "learning_rate": 6e-05,
      "loss": 2.8122,
      "step": 3
    },
    {
      "epoch": 0.0017472335468841002,
      "eval_loss": 2.335751533508301,
      "eval_runtime": 109.7628,
      "eval_samples_per_second": 6.587,
      "eval_steps_per_second": 3.298,
      "step": 3
    },
    {
      "epoch": 0.0023296447291788003,
      "grad_norm": 2.7917280197143555,
      "learning_rate": 8e-05,
      "loss": 2.1245,
      "step": 4
    },
    {
      "epoch": 0.0029120559114735,
      "grad_norm": 3.1003541946411133,
      "learning_rate": 0.0001,
      "loss": 2.2093,
      "step": 5
    },
    {
      "epoch": 0.0034944670937682005,
      "grad_norm": 3.3013908863067627,
      "learning_rate": 0.00012,
      "loss": 2.0963,
      "step": 6
    },
    {
      "epoch": 0.0034944670937682005,
      "eval_loss": 1.990991473197937,
      "eval_runtime": 110.0543,
      "eval_samples_per_second": 6.569,
      "eval_steps_per_second": 3.289,
      "step": 6
    },
    {
      "epoch": 0.004076878276062901,
      "grad_norm": 3.2542293071746826,
      "learning_rate": 0.00014,
      "loss": 2.2611,
      "step": 7
    },
    {
      "epoch": 0.004659289458357601,
      "grad_norm": 2.567934274673462,
      "learning_rate": 0.00016,
      "loss": 1.8188,
      "step": 8
    },
    {
      "epoch": 0.0052417006406523005,
      "grad_norm": 2.899068832397461,
      "learning_rate": 0.00018,
      "loss": 1.7527,
      "step": 9
    },
    {
      "epoch": 0.0052417006406523005,
      "eval_loss": 1.6364027261734009,
      "eval_runtime": 109.9246,
      "eval_samples_per_second": 6.577,
      "eval_steps_per_second": 3.293,
      "step": 9
    },
    {
      "epoch": 0.005824111822947,
      "grad_norm": 2.303987979888916,
      "learning_rate": 0.0002,
      "loss": 1.3736,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1935445139128320.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}