last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.021727322107550243,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0021727322107550242,
      "grad_norm": 37.16297912597656,
      "learning_rate": 2e-05,
      "loss": 26.2925,
      "step": 1
    },
    {
      "epoch": 0.0021727322107550242,
      "eval_loss": 6.362123489379883,
      "eval_runtime": 69.8308,
      "eval_samples_per_second": 2.778,
      "eval_steps_per_second": 1.389,
      "step": 1
    },
    {
      "epoch": 0.0043454644215100485,
      "grad_norm": 32.820220947265625,
      "learning_rate": 4e-05,
      "loss": 23.3629,
      "step": 2
    },
    {
      "epoch": 0.006518196632265073,
      "grad_norm": 45.58674621582031,
      "learning_rate": 6e-05,
      "loss": 31.4505,
      "step": 3
    },
    {
      "epoch": 0.006518196632265073,
      "eval_loss": 6.055120944976807,
      "eval_runtime": 68.8054,
      "eval_samples_per_second": 2.82,
      "eval_steps_per_second": 1.41,
      "step": 3
    },
    {
      "epoch": 0.008690928843020097,
      "grad_norm": 33.71705627441406,
      "learning_rate": 8e-05,
      "loss": 22.7458,
      "step": 4
    },
    {
      "epoch": 0.010863661053775122,
      "grad_norm": 26.417665481567383,
      "learning_rate": 0.0001,
      "loss": 19.5417,
      "step": 5
    },
    {
      "epoch": 0.013036393264530146,
      "grad_norm": 31.988718032836914,
      "learning_rate": 0.00012,
      "loss": 24.3794,
      "step": 6
    },
    {
      "epoch": 0.013036393264530146,
      "eval_loss": 4.7808332443237305,
      "eval_runtime": 67.3901,
      "eval_samples_per_second": 2.879,
      "eval_steps_per_second": 1.439,
      "step": 6
    },
    {
      "epoch": 0.015209125475285171,
      "grad_norm": 29.322267532348633,
      "learning_rate": 0.00014,
      "loss": 19.7093,
      "step": 7
    },
    {
      "epoch": 0.017381857686040194,
      "grad_norm": 27.37926483154297,
      "learning_rate": 0.00016,
      "loss": 18.7525,
      "step": 8
    },
    {
      "epoch": 0.01955458989679522,
      "grad_norm": 28.797134399414062,
      "learning_rate": 0.00018,
      "loss": 17.7134,
      "step": 9
    },
    {
      "epoch": 0.01955458989679522,
      "eval_loss": 3.6513190269470215,
      "eval_runtime": 68.1618,
      "eval_samples_per_second": 2.846,
      "eval_steps_per_second": 1.423,
      "step": 9
    },
    {
      "epoch": 0.021727322107550243,
      "grad_norm": 34.8036003112793,
      "learning_rate": 0.0002,
      "loss": 16.0472,
      "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": 1803185647779840.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}