last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0012322479282831705,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00012322479282831706,
      "grad_norm": 11.081421852111816,
      "learning_rate": 2e-05,
      "loss": 6.768,
      "step": 1
    },
    {
      "epoch": 0.00012322479282831706,
      "eval_loss": 1.7490488290786743,
      "eval_runtime": 165.9536,
      "eval_samples_per_second": 20.59,
      "eval_steps_per_second": 10.298,
      "step": 1
    },
    {
      "epoch": 0.0002464495856566341,
      "grad_norm": 10.705796241760254,
      "learning_rate": 4e-05,
      "loss": 7.1368,
      "step": 2
    },
    {
      "epoch": 0.0003696743784849512,
      "grad_norm": 9.518936157226562,
      "learning_rate": 6e-05,
      "loss": 6.9495,
      "step": 3
    },
    {
      "epoch": 0.0003696743784849512,
      "eval_loss": 1.7327848672866821,
      "eval_runtime": 164.4482,
      "eval_samples_per_second": 20.779,
      "eval_steps_per_second": 10.392,
      "step": 3
    },
    {
      "epoch": 0.0004928991713132682,
      "grad_norm": 9.401508331298828,
      "learning_rate": 8e-05,
      "loss": 7.1067,
      "step": 4
    },
    {
      "epoch": 0.0006161239641415853,
      "grad_norm": 9.68394947052002,
      "learning_rate": 0.0001,
      "loss": 7.2791,
      "step": 5
    },
    {
      "epoch": 0.0007393487569699024,
      "grad_norm": 9.809041023254395,
      "learning_rate": 0.00012,
      "loss": 7.1385,
      "step": 6
    },
    {
      "epoch": 0.0007393487569699024,
      "eval_loss": 1.6273040771484375,
      "eval_runtime": 164.4829,
      "eval_samples_per_second": 20.774,
      "eval_steps_per_second": 10.39,
      "step": 6
    },
    {
      "epoch": 0.0008625735497982194,
      "grad_norm": 8.082130432128906,
      "learning_rate": 0.00014,
      "loss": 6.4543,
      "step": 7
    },
    {
      "epoch": 0.0009857983426265365,
      "grad_norm": 6.84332799911499,
      "learning_rate": 0.00016,
      "loss": 5.9262,
      "step": 8
    },
    {
      "epoch": 0.0011090231354548535,
      "grad_norm": 8.637106895446777,
      "learning_rate": 0.00018,
      "loss": 6.549,
      "step": 9
    },
    {
      "epoch": 0.0011090231354548535,
      "eval_loss": 1.6292555332183838,
      "eval_runtime": 164.3875,
      "eval_samples_per_second": 20.786,
      "eval_steps_per_second": 10.396,
      "step": 9
    },
    {
      "epoch": 0.0012322479282831705,
      "grad_norm": 6.6092023849487305,
      "learning_rate": 0.0002,
      "loss": 6.0184,
      "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": 2891908468703232.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}