last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0007492180037086292,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 7.492180037086291e-05,
      "grad_norm": 43.05109786987305,
      "learning_rate": 2e-05,
      "loss": 4.2125,
      "step": 1
    },
    {
      "epoch": 7.492180037086291e-05,
      "eval_loss": 4.021458148956299,
      "eval_runtime": 83.7736,
      "eval_samples_per_second": 67.086,
      "eval_steps_per_second": 33.543,
      "step": 1
    },
    {
      "epoch": 0.00014984360074172582,
      "grad_norm": 39.96201705932617,
      "learning_rate": 4e-05,
      "loss": 2.5337,
      "step": 2
    },
    {
      "epoch": 0.00022476540111258873,
      "grad_norm": 34.15665054321289,
      "learning_rate": 6e-05,
      "loss": 3.0091,
      "step": 3
    },
    {
      "epoch": 0.00022476540111258873,
      "eval_loss": 3.7696897983551025,
      "eval_runtime": 82.9136,
      "eval_samples_per_second": 67.781,
      "eval_steps_per_second": 33.891,
      "step": 3
    },
    {
      "epoch": 0.00029968720148345165,
      "grad_norm": 13.744331359863281,
      "learning_rate": 8e-05,
      "loss": 2.173,
      "step": 4
    },
    {
      "epoch": 0.0003746090018543146,
      "grad_norm": 7.830999851226807,
      "learning_rate": 0.0001,
      "loss": 2.0188,
      "step": 5
    },
    {
      "epoch": 0.00044953080222517747,
      "grad_norm": 24.29507064819336,
      "learning_rate": 0.00012,
      "loss": 3.5455,
      "step": 6
    },
    {
      "epoch": 0.00044953080222517747,
      "eval_loss": 2.771317481994629,
      "eval_runtime": 80.9009,
      "eval_samples_per_second": 69.468,
      "eval_steps_per_second": 34.734,
      "step": 6
    },
    {
      "epoch": 0.0005244526025960404,
      "grad_norm": 6.703357219696045,
      "learning_rate": 0.00014,
      "loss": 1.8121,
      "step": 7
    },
    {
      "epoch": 0.0005993744029669033,
      "grad_norm": 6.9556427001953125,
      "learning_rate": 0.00016,
      "loss": 2.5551,
      "step": 8
    },
    {
      "epoch": 0.0006742962033377662,
      "grad_norm": 8.209716796875,
      "learning_rate": 0.00018,
      "loss": 2.8194,
      "step": 9
    },
    {
      "epoch": 0.0006742962033377662,
      "eval_loss": 1.9984983205795288,
      "eval_runtime": 81.9238,
      "eval_samples_per_second": 68.6,
      "eval_steps_per_second": 34.3,
      "step": 9
    },
    {
      "epoch": 0.0007492180037086292,
      "grad_norm": 16.12192726135254,
      "learning_rate": 0.0002,
      "loss": 3.0209,
      "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": 255791978250240.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}