last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00831255195344971,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008312551953449709,
      "grad_norm": 2.9409139156341553,
      "learning_rate": 2e-05,
      "loss": 10.8636,
      "step": 1
    },
    {
      "epoch": 0.0008312551953449709,
      "eval_loss": 2.586120128631592,
      "eval_runtime": 31.2605,
      "eval_samples_per_second": 16.219,
      "eval_steps_per_second": 8.125,
      "step": 1
    },
    {
      "epoch": 0.0016625103906899418,
      "grad_norm": 1.0974133014678955,
      "learning_rate": 4e-05,
      "loss": 9.5429,
      "step": 2
    },
    {
      "epoch": 0.0024937655860349127,
      "grad_norm": 1.2267212867736816,
      "learning_rate": 6e-05,
      "loss": 8.0152,
      "step": 3
    },
    {
      "epoch": 0.0024937655860349127,
      "eval_loss": 2.584916830062866,
      "eval_runtime": 31.2786,
      "eval_samples_per_second": 16.209,
      "eval_steps_per_second": 8.121,
      "step": 3
    },
    {
      "epoch": 0.0033250207813798837,
      "grad_norm": 0.8856640458106995,
      "learning_rate": 8e-05,
      "loss": 7.8662,
      "step": 4
    },
    {
      "epoch": 0.004156275976724855,
      "grad_norm": 1.3124897480010986,
      "learning_rate": 0.0001,
      "loss": 7.6642,
      "step": 5
    },
    {
      "epoch": 0.004987531172069825,
      "grad_norm": 0.9891735315322876,
      "learning_rate": 0.00012,
      "loss": 6.5274,
      "step": 6
    },
    {
      "epoch": 0.004987531172069825,
      "eval_loss": 2.5601935386657715,
      "eval_runtime": 31.2751,
      "eval_samples_per_second": 16.211,
      "eval_steps_per_second": 8.121,
      "step": 6
    },
    {
      "epoch": 0.005818786367414797,
      "grad_norm": 1.226442575454712,
      "learning_rate": 0.00014,
      "loss": 9.9632,
      "step": 7
    },
    {
      "epoch": 0.006650041562759767,
      "grad_norm": 1.5548657178878784,
      "learning_rate": 0.00016,
      "loss": 10.1128,
      "step": 8
    },
    {
      "epoch": 0.007481296758104738,
      "grad_norm": 1.2843385934829712,
      "learning_rate": 0.00018,
      "loss": 7.83,
      "step": 9
    },
    {
      "epoch": 0.007481296758104738,
      "eval_loss": 2.434023857116699,
      "eval_runtime": 31.307,
      "eval_samples_per_second": 16.194,
      "eval_steps_per_second": 8.113,
      "step": 9
    },
    {
      "epoch": 0.00831255195344971,
      "grad_norm": 2.356943368911743,
      "learning_rate": 0.0002,
      "loss": 8.5216,
      "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": 3562002343526400.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}