last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.03736007142366596,
  "eval_steps": 50,
  "global_step": 34,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0010988256301078224,
      "grad_norm": NaN,
      "learning_rate": 7.499999999999999e-06,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.0010988256301078224,
      "eval_loss": NaN,
      "eval_runtime": 586.4445,
      "eval_samples_per_second": 5.228,
      "eval_steps_per_second": 2.614,
      "step": 1
    },
    {
      "epoch": 0.0021976512602156447,
      "grad_norm": NaN,
      "learning_rate": 1.4999999999999999e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.003296476890323467,
      "grad_norm": NaN,
      "learning_rate": 2.2499999999999998e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.0043953025204312894,
      "grad_norm": NaN,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.005494128150539112,
      "grad_norm": NaN,
      "learning_rate": 3.75e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.006592953780646934,
      "grad_norm": NaN,
      "learning_rate": 4.4999999999999996e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.007691779410754756,
      "grad_norm": NaN,
      "learning_rate": 5.2499999999999995e-05,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.008790605040862579,
      "grad_norm": NaN,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.0098894306709704,
      "grad_norm": NaN,
      "learning_rate": 6.75e-05,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.010988256301078223,
      "grad_norm": NaN,
      "learning_rate": 7.5e-05,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.012087081931186044,
      "grad_norm": NaN,
      "learning_rate": 8.25e-05,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.013185907561293867,
      "grad_norm": NaN,
      "learning_rate": 8.999999999999999e-05,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.014284733191401689,
      "grad_norm": NaN,
      "learning_rate": 9.75e-05,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.015383558821509512,
      "grad_norm": NaN,
      "learning_rate": 0.00010499999999999999,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.016482384451617333,
      "grad_norm": NaN,
      "learning_rate": 0.0001125,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.017581210081725158,
      "grad_norm": NaN,
      "learning_rate": 0.00011999999999999999,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 0.01868003571183298,
      "grad_norm": NaN,
      "learning_rate": 0.00012749999999999998,
      "loss": 0.0,
      "step": 17
    },
    {
      "epoch": 0.0197788613419408,
      "grad_norm": NaN,
      "learning_rate": 0.000135,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.02087768697204862,
      "grad_norm": NaN,
      "learning_rate": 0.0001425,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 0.021976512602156446,
      "grad_norm": NaN,
      "learning_rate": 0.00015,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.023075338232264268,
      "grad_norm": NaN,
      "learning_rate": 0.00014998857713672935,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.02417416386237209,
      "grad_norm": NaN,
      "learning_rate": 0.00014995431202643217,
      "loss": 0.0,
      "step": 22
    },
    {
      "epoch": 0.025272989492479914,
      "grad_norm": NaN,
      "learning_rate": 0.000149897215106593,
      "loss": 0.0,
      "step": 23
    },
    {
      "epoch": 0.026371815122587735,
      "grad_norm": NaN,
      "learning_rate": 0.0001498173037694868,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.027470640752695556,
      "grad_norm": NaN,
      "learning_rate": 0.0001497146023568809,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.028569466382803377,
      "grad_norm": NaN,
      "learning_rate": 0.00014958914215262048,
      "loss": 0.0,
      "step": 26
    },
    {
      "epoch": 0.029668292012911202,
      "grad_norm": NaN,
      "learning_rate": 0.00014944096137309914,
      "loss": 0.0,
      "step": 27
    },
    {
      "epoch": 0.030767117643019024,
      "grad_norm": NaN,
      "learning_rate": 0.00014927010515561776,
      "loss": 0.0,
      "step": 28
    },
    {
      "epoch": 0.03186594327312685,
      "grad_norm": NaN,
      "learning_rate": 0.00014907662554463532,
      "loss": 0.0,
      "step": 29
    },
    {
      "epoch": 0.032964768903234666,
      "grad_norm": NaN,
      "learning_rate": 0.0001488605814759156,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.03406359453334249,
      "grad_norm": NaN,
      "learning_rate": 0.00014862203875857477,
      "loss": 0.0,
      "step": 31
    },
    {
      "epoch": 0.035162420163450316,
      "grad_norm": NaN,
      "learning_rate": 0.0001483610700550354,
      "loss": 0.0,
      "step": 32
    },
    {
      "epoch": 0.03626124579355813,
      "grad_norm": NaN,
      "learning_rate": 0.00014807775485889264,
      "loss": 0.0,
      "step": 33
    },
    {
      "epoch": 0.03736007142366596,
      "grad_norm": NaN,
      "learning_rate": 0.0001477721794706997,
      "loss": 0.0,
      "step": 34
    }
  ],
  "logging_steps": 1,
  "max_steps": 200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 17,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.0001228862154342e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}