last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00024782379728013383,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.478237972801338e-05,
      "grad_norm": 0.5899750590324402,
      "learning_rate": 2e-05,
      "loss": 0.6861,
      "step": 1
    },
    {
      "epoch": 2.478237972801338e-05,
      "eval_loss": 0.9835758209228516,
      "eval_runtime": 628.0838,
      "eval_samples_per_second": 27.051,
      "eval_steps_per_second": 13.525,
      "step": 1
    },
    {
      "epoch": 4.956475945602676e-05,
      "grad_norm": 0.9787094593048096,
      "learning_rate": 4e-05,
      "loss": 0.8777,
      "step": 2
    },
    {
      "epoch": 7.434713918404015e-05,
      "grad_norm": 0.980866551399231,
      "learning_rate": 6e-05,
      "loss": 0.9359,
      "step": 3
    },
    {
      "epoch": 7.434713918404015e-05,
      "eval_loss": 0.9763832688331604,
      "eval_runtime": 625.7647,
      "eval_samples_per_second": 27.151,
      "eval_steps_per_second": 13.575,
      "step": 3
    },
    {
      "epoch": 9.912951891205352e-05,
      "grad_norm": 0.9241377711296082,
      "learning_rate": 8e-05,
      "loss": 1.0659,
      "step": 4
    },
    {
      "epoch": 0.00012391189864006691,
      "grad_norm": 0.7938036918640137,
      "learning_rate": 0.0001,
      "loss": 0.8137,
      "step": 5
    },
    {
      "epoch": 0.0001486942783680803,
      "grad_norm": 1.0245213508605957,
      "learning_rate": 0.00012,
      "loss": 1.1537,
      "step": 6
    },
    {
      "epoch": 0.0001486942783680803,
      "eval_loss": 0.8665831685066223,
      "eval_runtime": 627.8054,
      "eval_samples_per_second": 27.063,
      "eval_steps_per_second": 13.531,
      "step": 6
    },
    {
      "epoch": 0.00017347665809609367,
      "grad_norm": 0.6237964630126953,
      "learning_rate": 0.00014,
      "loss": 1.0758,
      "step": 7
    },
    {
      "epoch": 0.00019825903782410705,
      "grad_norm": 0.6559966206550598,
      "learning_rate": 0.00016,
      "loss": 0.909,
      "step": 8
    },
    {
      "epoch": 0.00022304141755212045,
      "grad_norm": 0.6187783479690552,
      "learning_rate": 0.00018,
      "loss": 0.7102,
      "step": 9
    },
    {
      "epoch": 0.00022304141755212045,
      "eval_loss": 0.7870052456855774,
      "eval_runtime": 628.2559,
      "eval_samples_per_second": 27.043,
      "eval_steps_per_second": 13.522,
      "step": 9
    },
    {
      "epoch": 0.00024782379728013383,
      "grad_norm": 0.5213778018951416,
      "learning_rate": 0.0002,
      "loss": 0.6475,
      "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": 2265716203978752.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}