last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.023653088042049936,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002628120893561104,
      "grad_norm": 1.3634989261627197,
      "learning_rate": 2e-05,
      "loss": 3.5272,
      "step": 1
    },
    {
      "epoch": 0.002628120893561104,
      "eval_loss": 3.377916097640991,
      "eval_runtime": 11.2615,
      "eval_samples_per_second": 14.296,
      "eval_steps_per_second": 7.193,
      "step": 1
    },
    {
      "epoch": 0.005256241787122208,
      "grad_norm": 1.4552209377288818,
      "learning_rate": 4e-05,
      "loss": 3.5758,
      "step": 2
    },
    {
      "epoch": 0.00788436268068331,
      "grad_norm": 1.2623372077941895,
      "learning_rate": 6e-05,
      "loss": 3.3248,
      "step": 3
    },
    {
      "epoch": 0.00788436268068331,
      "eval_loss": 3.3656370639801025,
      "eval_runtime": 11.1958,
      "eval_samples_per_second": 14.38,
      "eval_steps_per_second": 7.235,
      "step": 3
    },
    {
      "epoch": 0.010512483574244415,
      "grad_norm": 1.6191483736038208,
      "learning_rate": 8e-05,
      "loss": 3.4872,
      "step": 4
    },
    {
      "epoch": 0.013140604467805518,
      "grad_norm": 1.4146218299865723,
      "learning_rate": 0.0001,
      "loss": 3.1748,
      "step": 5
    },
    {
      "epoch": 0.01576872536136662,
      "grad_norm": 1.7842243909835815,
      "learning_rate": 0.00012,
      "loss": 3.2239,
      "step": 6
    },
    {
      "epoch": 0.01576872536136662,
      "eval_loss": 3.25826358795166,
      "eval_runtime": 11.3071,
      "eval_samples_per_second": 14.239,
      "eval_steps_per_second": 7.164,
      "step": 6
    },
    {
      "epoch": 0.018396846254927726,
      "grad_norm": 1.9601093530654907,
      "learning_rate": 0.00014,
      "loss": 3.2528,
      "step": 7
    },
    {
      "epoch": 0.02102496714848883,
      "grad_norm": 1.72002375125885,
      "learning_rate": 0.00016,
      "loss": 2.8813,
      "step": 8
    },
    {
      "epoch": 0.023653088042049936,
      "grad_norm": 1.4412205219268799,
      "learning_rate": 0.00018,
      "loss": 2.7633,
      "step": 9
    },
    {
      "epoch": 0.023653088042049936,
      "eval_loss": 3.0437607765197754,
      "eval_runtime": 11.297,
      "eval_samples_per_second": 14.252,
      "eval_steps_per_second": 7.17,
      "step": 9
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3329215646662656.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}