last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004968326915911067,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0004968326915911067,
      "grad_norm": 8.483444213867188,
      "learning_rate": 2e-05,
      "loss": 13.4967,
      "step": 1
    },
    {
      "epoch": 0.0004968326915911067,
      "eval_loss": 4.478808879852295,
      "eval_runtime": 9.8305,
      "eval_samples_per_second": 86.262,
      "eval_steps_per_second": 43.131,
      "step": 1
    },
    {
      "epoch": 0.0009936653831822133,
      "grad_norm": 13.397222518920898,
      "learning_rate": 4e-05,
      "loss": 18.8801,
      "step": 2
    },
    {
      "epoch": 0.0014904980747733202,
      "grad_norm": 10.576736450195312,
      "learning_rate": 6e-05,
      "loss": 15.9005,
      "step": 3
    },
    {
      "epoch": 0.0014904980747733202,
      "eval_loss": 4.446848392486572,
      "eval_runtime": 9.5204,
      "eval_samples_per_second": 89.072,
      "eval_steps_per_second": 44.536,
      "step": 3
    },
    {
      "epoch": 0.0019873307663644266,
      "grad_norm": 13.7582368850708,
      "learning_rate": 8e-05,
      "loss": 19.2427,
      "step": 4
    },
    {
      "epoch": 0.0024841634579555335,
      "grad_norm": 10.629317283630371,
      "learning_rate": 0.0001,
      "loss": 13.8256,
      "step": 5
    },
    {
      "epoch": 0.0029809961495466403,
      "grad_norm": 11.852766036987305,
      "learning_rate": 0.00012,
      "loss": 18.2366,
      "step": 6
    },
    {
      "epoch": 0.0029809961495466403,
      "eval_loss": 4.262442111968994,
      "eval_runtime": 9.5673,
      "eval_samples_per_second": 88.635,
      "eval_steps_per_second": 44.317,
      "step": 6
    },
    {
      "epoch": 0.0034778288411377468,
      "grad_norm": 9.952699661254883,
      "learning_rate": 0.00014,
      "loss": 16.266,
      "step": 7
    },
    {
      "epoch": 0.003974661532728853,
      "grad_norm": 17.484481811523438,
      "learning_rate": 0.00016,
      "loss": 20.548,
      "step": 8
    },
    {
      "epoch": 0.0044714942243199605,
      "grad_norm": 17.841690063476562,
      "learning_rate": 0.00018,
      "loss": 18.8747,
      "step": 9
    },
    {
      "epoch": 0.0044714942243199605,
      "eval_loss": 3.8306868076324463,
      "eval_runtime": 9.6738,
      "eval_samples_per_second": 87.66,
      "eval_steps_per_second": 43.83,
      "step": 9
    },
    {
      "epoch": 0.004968326915911067,
      "grad_norm": 12.73868465423584,
      "learning_rate": 0.0002,
      "loss": 15.9641,
      "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": 75069652992000.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}