last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.01002004008016032,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.001002004008016032,
      "grad_norm": 0.4462810456752777,
      "learning_rate": 2e-05,
      "loss": 6.083,
      "step": 1
    },
    {
      "epoch": 0.001002004008016032,
      "eval_loss": 1.4782618284225464,
      "eval_runtime": 38.6879,
      "eval_samples_per_second": 10.882,
      "eval_steps_per_second": 5.454,
      "step": 1
    },
    {
      "epoch": 0.002004008016032064,
      "grad_norm": 0.5297453999519348,
      "learning_rate": 4e-05,
      "loss": 6.1034,
      "step": 2
    },
    {
      "epoch": 0.003006012024048096,
      "grad_norm": 0.45829904079437256,
      "learning_rate": 6e-05,
      "loss": 6.2999,
      "step": 3
    },
    {
      "epoch": 0.003006012024048096,
      "eval_loss": 1.4774115085601807,
      "eval_runtime": 38.7104,
      "eval_samples_per_second": 10.876,
      "eval_steps_per_second": 5.451,
      "step": 3
    },
    {
      "epoch": 0.004008016032064128,
      "grad_norm": 0.43833157420158386,
      "learning_rate": 8e-05,
      "loss": 5.5947,
      "step": 4
    },
    {
      "epoch": 0.00501002004008016,
      "grad_norm": 0.500750720500946,
      "learning_rate": 0.0001,
      "loss": 5.5166,
      "step": 5
    },
    {
      "epoch": 0.006012024048096192,
      "grad_norm": 0.5694366097450256,
      "learning_rate": 0.00012,
      "loss": 5.6152,
      "step": 6
    },
    {
      "epoch": 0.006012024048096192,
      "eval_loss": 1.4591740369796753,
      "eval_runtime": 38.8132,
      "eval_samples_per_second": 10.847,
      "eval_steps_per_second": 5.436,
      "step": 6
    },
    {
      "epoch": 0.0070140280561122245,
      "grad_norm": 0.6731088161468506,
      "learning_rate": 0.00014,
      "loss": 5.417,
      "step": 7
    },
    {
      "epoch": 0.008016032064128256,
      "grad_norm": 0.7657998204231262,
      "learning_rate": 0.00016,
      "loss": 5.9049,
      "step": 8
    },
    {
      "epoch": 0.009018036072144289,
      "grad_norm": 1.130814790725708,
      "learning_rate": 0.00018,
      "loss": 5.6898,
      "step": 9
    },
    {
      "epoch": 0.009018036072144289,
      "eval_loss": 1.3330568075180054,
      "eval_runtime": 38.7627,
      "eval_samples_per_second": 10.861,
      "eval_steps_per_second": 5.443,
      "step": 9
    },
    {
      "epoch": 0.01002004008016032,
      "grad_norm": 1.1171002388000488,
      "learning_rate": 0.0002,
      "loss": 5.4603,
      "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": 6332448610713600.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}