checkpoint-12/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.96,
  "eval_steps": 3,
  "global_step": 12,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "grad_norm": 0.1656520962715149,
      "learning_rate": 2e-05,
      "loss": 1.4615,
      "step": 1
    },
    {
      "epoch": 0.08,
      "eval_loss": 1.4899382591247559,
      "eval_runtime": 29.186,
      "eval_samples_per_second": 3.426,
      "eval_steps_per_second": 1.713,
      "step": 1
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.1882542371749878,
      "learning_rate": 4e-05,
      "loss": 1.4241,
      "step": 2
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.15945176780223846,
      "learning_rate": 6e-05,
      "loss": 1.3849,
      "step": 3
    },
    {
      "epoch": 0.24,
      "eval_loss": 1.485183835029602,
      "eval_runtime": 29.3484,
      "eval_samples_per_second": 3.407,
      "eval_steps_per_second": 1.704,
      "step": 3
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.13675835728645325,
      "learning_rate": 8e-05,
      "loss": 1.2212,
      "step": 4
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.1532098948955536,
      "learning_rate": 0.0001,
      "loss": 1.3626,
      "step": 5
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.16159594058990479,
      "learning_rate": 0.00012,
      "loss": 1.3665,
      "step": 6
    },
    {
      "epoch": 0.48,
      "eval_loss": 1.441084623336792,
      "eval_runtime": 29.4785,
      "eval_samples_per_second": 3.392,
      "eval_steps_per_second": 1.696,
      "step": 6
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.1462002545595169,
      "learning_rate": 0.00014,
      "loss": 1.3003,
      "step": 7
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.13418763875961304,
      "learning_rate": 0.00016,
      "loss": 1.3331,
      "step": 8
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.10984567552804947,
      "learning_rate": 0.00018,
      "loss": 1.2689,
      "step": 9
    },
    {
      "epoch": 0.72,
      "eval_loss": 1.3380621671676636,
      "eval_runtime": 29.5305,
      "eval_samples_per_second": 3.386,
      "eval_steps_per_second": 1.693,
      "step": 9
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.10075916349887848,
      "learning_rate": 0.0002,
      "loss": 1.2942,
      "step": 10
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.11774784326553345,
      "learning_rate": 0.000199658449300667,
      "loss": 1.2936,
      "step": 11
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.10691336542367935,
      "learning_rate": 0.00019863613034027224,
      "loss": 1.2258,
      "step": 12
    },
    {
      "epoch": 0.96,
      "eval_loss": 1.296021580696106,
      "eval_runtime": 29.6291,
      "eval_samples_per_second": 3.375,
      "eval_steps_per_second": 1.688,
      "step": 12
    }
  ],
  "logging_steps": 1,
  "max_steps": 48,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 12,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2500249176244224.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}