last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.014847809948032665,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0014847809948032665,
      "grad_norm": 5.69954252243042,
      "learning_rate": 2e-05,
      "loss": 11.3933,
      "step": 1
    },
    {
      "epoch": 0.002969561989606533,
      "grad_norm": 4.038008689880371,
      "learning_rate": 4e-05,
      "loss": 11.0483,
      "step": 2
    },
    {
      "epoch": 0.004454342984409799,
      "grad_norm": 4.3309783935546875,
      "learning_rate": 6e-05,
      "loss": 13.5248,
      "step": 3
    },
    {
      "epoch": 0.004454342984409799,
      "eval_loss": 2.832982063293457,
      "eval_runtime": 15.8956,
      "eval_samples_per_second": 17.867,
      "eval_steps_per_second": 8.933,
      "step": 3
    },
    {
      "epoch": 0.005939123979213066,
      "grad_norm": 3.093883514404297,
      "learning_rate": 8e-05,
      "loss": 8.7892,
      "step": 4
    },
    {
      "epoch": 0.007423904974016332,
      "grad_norm": 6.267045497894287,
      "learning_rate": 0.0001,
      "loss": 13.5236,
      "step": 5
    },
    {
      "epoch": 0.008908685968819599,
      "grad_norm": 8.435811996459961,
      "learning_rate": 0.00012,
      "loss": 12.1843,
      "step": 6
    },
    {
      "epoch": 0.008908685968819599,
      "eval_loss": 2.572983741760254,
      "eval_runtime": 20.6336,
      "eval_samples_per_second": 13.764,
      "eval_steps_per_second": 6.882,
      "step": 6
    },
    {
      "epoch": 0.010393466963622866,
      "grad_norm": 10.377184867858887,
      "learning_rate": 0.00014,
      "loss": 9.6,
      "step": 7
    },
    {
      "epoch": 0.011878247958426132,
      "grad_norm": 21.33673667907715,
      "learning_rate": 0.00016,
      "loss": 9.8471,
      "step": 8
    },
    {
      "epoch": 0.013363028953229399,
      "grad_norm": 15.13981819152832,
      "learning_rate": 0.00018,
      "loss": 6.9279,
      "step": 9
    },
    {
      "epoch": 0.013363028953229399,
      "eval_loss": 1.2549283504486084,
      "eval_runtime": 18.303,
      "eval_samples_per_second": 15.517,
      "eval_steps_per_second": 7.758,
      "step": 9
    },
    {
      "epoch": 0.014847809948032665,
      "grad_norm": 5.209914207458496,
      "learning_rate": 0.0002,
      "loss": 4.6903,
      "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": 3166224305356800.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}