last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0007662394881520219,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 7.662394881520219e-05,
      "grad_norm": 2.090266704559326,
      "learning_rate": 2e-05,
      "loss": 0.8768,
      "step": 1
    },
    {
      "epoch": 7.662394881520219e-05,
      "eval_loss": 0.8993696570396423,
      "eval_runtime": 1496.3373,
      "eval_samples_per_second": 3.672,
      "eval_steps_per_second": 1.836,
      "step": 1
    },
    {
      "epoch": 0.00015324789763040438,
      "grad_norm": 2.377678155899048,
      "learning_rate": 4e-05,
      "loss": 0.8815,
      "step": 2
    },
    {
      "epoch": 0.00022987184644560657,
      "grad_norm": 2.19423246383667,
      "learning_rate": 6e-05,
      "loss": 0.8734,
      "step": 3
    },
    {
      "epoch": 0.00022987184644560657,
      "eval_loss": 0.8186013102531433,
      "eval_runtime": 1500.6509,
      "eval_samples_per_second": 3.662,
      "eval_steps_per_second": 1.831,
      "step": 3
    },
    {
      "epoch": 0.00030649579526080876,
      "grad_norm": 2.075087070465088,
      "learning_rate": 8e-05,
      "loss": 0.8706,
      "step": 4
    },
    {
      "epoch": 0.00038311974407601095,
      "grad_norm": 1.9129090309143066,
      "learning_rate": 0.0001,
      "loss": 0.6452,
      "step": 5
    },
    {
      "epoch": 0.00045974369289121313,
      "grad_norm": 1.933599591255188,
      "learning_rate": 0.00012,
      "loss": 0.4879,
      "step": 6
    },
    {
      "epoch": 0.00045974369289121313,
      "eval_loss": 0.34602582454681396,
      "eval_runtime": 1500.8477,
      "eval_samples_per_second": 3.661,
      "eval_steps_per_second": 1.831,
      "step": 6
    },
    {
      "epoch": 0.0005363676417064154,
      "grad_norm": 1.934218406677246,
      "learning_rate": 0.00014,
      "loss": 0.3667,
      "step": 7
    },
    {
      "epoch": 0.0006129915905216175,
      "grad_norm": 1.2223925590515137,
      "learning_rate": 0.00016,
      "loss": 0.2294,
      "step": 8
    },
    {
      "epoch": 0.0006896155393368198,
      "grad_norm": 1.2483911514282227,
      "learning_rate": 0.00018,
      "loss": 0.136,
      "step": 9
    },
    {
      "epoch": 0.0006896155393368198,
      "eval_loss": 0.08404787629842758,
      "eval_runtime": 1502.7204,
      "eval_samples_per_second": 3.657,
      "eval_steps_per_second": 1.829,
      "step": 9
    },
    {
      "epoch": 0.0007662394881520219,
      "grad_norm": 0.6040265560150146,
      "learning_rate": 0.0002,
      "loss": 0.083,
      "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": 6585384016281600.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}