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": 0.17093713581562042,
      "learning_rate": 2e-05,
      "loss": 1.4615,
      "step": 1
    },
    {
      "epoch": 0.08,
      "eval_loss": 1.4899382591247559,
      "eval_runtime": 17.3107,
      "eval_samples_per_second": 5.777,
      "eval_steps_per_second": 2.888,
      "step": 1
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.19339510798454285,
      "learning_rate": 4e-05,
      "loss": 1.4241,
      "step": 2
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.1669788658618927,
      "learning_rate": 6e-05,
      "loss": 1.3846,
      "step": 3
    },
    {
      "epoch": 0.24,
      "eval_loss": 1.4858685731887817,
      "eval_runtime": 17.4215,
      "eval_samples_per_second": 5.74,
      "eval_steps_per_second": 2.87,
      "step": 3
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.14142441749572754,
      "learning_rate": 8e-05,
      "loss": 1.2219,
      "step": 4
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.15717843174934387,
      "learning_rate": 0.0001,
      "loss": 1.3617,
      "step": 5
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.16309261322021484,
      "learning_rate": 0.00012,
      "loss": 1.3667,
      "step": 6
    },
    {
      "epoch": 0.48,
      "eval_loss": 1.439871072769165,
      "eval_runtime": 17.6339,
      "eval_samples_per_second": 5.671,
      "eval_steps_per_second": 2.835,
      "step": 6
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.15078029036521912,
      "learning_rate": 0.00014,
      "loss": 1.3008,
      "step": 7
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.13603582978248596,
      "learning_rate": 0.00016,
      "loss": 1.3333,
      "step": 8
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.1104956567287445,
      "learning_rate": 0.00018,
      "loss": 1.267,
      "step": 9
    },
    {
      "epoch": 0.72,
      "eval_loss": 1.3381670713424683,
      "eval_runtime": 17.2986,
      "eval_samples_per_second": 5.781,
      "eval_steps_per_second": 2.89,
      "step": 9
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.09913735836744308,
      "learning_rate": 0.0002,
      "loss": 1.2946,
      "step": 10
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.11903145164251328,
      "learning_rate": 0.000199658449300667,
      "loss": 1.2921,
      "step": 11
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.11169299483299255,
      "learning_rate": 0.00019863613034027224,
      "loss": 1.2276,
      "step": 12
    },
    {
      "epoch": 0.96,
      "eval_loss": 1.2940881252288818,
      "eval_runtime": 17.4061,
      "eval_samples_per_second": 5.745,
      "eval_steps_per_second": 2.873,
      "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
}