last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.03473186996387886,
  "eval_steps": 63,
  "global_step": 250,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0001389274798555154,
      "eval_loss": 1.2610814571380615,
      "eval_runtime": 325.7738,
      "eval_samples_per_second": 9.304,
      "eval_steps_per_second": 4.654,
      "step": 1
    },
    {
      "epoch": 0.0013892747985551543,
      "grad_norm": 0.7473278045654297,
      "learning_rate": 0.00019979453927503364,
      "loss": 4.9082,
      "step": 10
    },
    {
      "epoch": 0.0027785495971103086,
      "grad_norm": 1.0333387851715088,
      "learning_rate": 0.00019815591569910654,
      "loss": 3.7899,
      "step": 20
    },
    {
      "epoch": 0.004167824395665463,
      "grad_norm": 0.8966497182846069,
      "learning_rate": 0.00019490557470106686,
      "loss": 4.3995,
      "step": 30
    },
    {
      "epoch": 0.005557099194220617,
      "grad_norm": 1.2714602947235107,
      "learning_rate": 0.0001900968867902419,
      "loss": 3.869,
      "step": 40
    },
    {
      "epoch": 0.006946373992775771,
      "grad_norm": 1.117074966430664,
      "learning_rate": 0.00018380881048918405,
      "loss": 3.839,
      "step": 50
    },
    {
      "epoch": 0.008335648791330925,
      "grad_norm": 0.7396908402442932,
      "learning_rate": 0.00017614459583691346,
      "loss": 3.9253,
      "step": 60
    },
    {
      "epoch": 0.008752431230897471,
      "eval_loss": 0.9566418528556824,
      "eval_runtime": 325.7857,
      "eval_samples_per_second": 9.304,
      "eval_steps_per_second": 4.653,
      "step": 63
    },
    {
      "epoch": 0.009724923589886079,
      "grad_norm": 0.9031943678855896,
      "learning_rate": 0.0001672300890261317,
      "loss": 3.7073,
      "step": 70
    },
    {
      "epoch": 0.011114198388441235,
      "grad_norm": 1.573042392730713,
      "learning_rate": 0.00015721166601221698,
      "loss": 3.8751,
      "step": 80
    },
    {
      "epoch": 0.012503473186996388,
      "grad_norm": 0.8371868133544922,
      "learning_rate": 0.00014625382902408356,
      "loss": 3.6648,
      "step": 90
    },
    {
      "epoch": 0.013892747985551542,
      "grad_norm": 1.2980339527130127,
      "learning_rate": 0.00013453650544213076,
      "loss": 3.6346,
      "step": 100
    },
    {
      "epoch": 0.015282022784106696,
      "grad_norm": 1.1150926351547241,
      "learning_rate": 0.00012225209339563145,
      "loss": 4.0115,
      "step": 110
    },
    {
      "epoch": 0.01667129758266185,
      "grad_norm": 2.019386053085327,
      "learning_rate": 0.00010960230259076818,
      "loss": 3.7301,
      "step": 120
    },
    {
      "epoch": 0.017504862461794942,
      "eval_loss": 0.9256976246833801,
      "eval_runtime": 325.9027,
      "eval_samples_per_second": 9.3,
      "eval_steps_per_second": 4.652,
      "step": 126
    },
    {
      "epoch": 0.018060572381217006,
      "grad_norm": 1.2206391096115112,
      "learning_rate": 9.679484224283449e-05,
      "loss": 3.9945,
      "step": 130
    },
    {
      "epoch": 0.019449847179772158,
      "grad_norm": 2.1848301887512207,
      "learning_rate": 8.404001049666211e-05,
      "loss": 4.1489,
      "step": 140
    },
    {
      "epoch": 0.020839121978327314,
      "grad_norm": 1.8375428915023804,
      "learning_rate": 7.154724133689677e-05,
      "loss": 3.6826,
      "step": 150
    },
    {
      "epoch": 0.02222839677688247,
      "grad_norm": 1.4407734870910645,
      "learning_rate": 5.952166568776062e-05,
      "loss": 3.4386,
      "step": 160
    },
    {
      "epoch": 0.02361767157543762,
      "grad_norm": 1.606547236442566,
      "learning_rate": 4.8160743168947496e-05,
      "loss": 3.7407,
      "step": 170
    },
    {
      "epoch": 0.025006946373992776,
      "grad_norm": 1.4281450510025024,
      "learning_rate": 3.7651019814126654e-05,
      "loss": 3.6141,
      "step": 180
    },
    {
      "epoch": 0.026257293692692415,
      "eval_loss": 0.9118097424507141,
      "eval_runtime": 325.9436,
      "eval_samples_per_second": 9.299,
      "eval_steps_per_second": 4.651,
      "step": 189
    },
    {
      "epoch": 0.02639622117254793,
      "grad_norm": 1.2719013690948486,
      "learning_rate": 2.8165064990227252e-05,
      "loss": 3.3775,
      "step": 190
    },
    {
      "epoch": 0.027785495971103084,
      "grad_norm": 1.359384536743164,
      "learning_rate": 1.985863781320435e-05,
      "loss": 3.3751,
      "step": 200
    },
    {
      "epoch": 0.02917477076965824,
      "grad_norm": 1.3881595134735107,
      "learning_rate": 1.286812958766106e-05,
      "loss": 3.7558,
      "step": 210
    },
    {
      "epoch": 0.03056404556821339,
      "grad_norm": 1.2139527797698975,
      "learning_rate": 7.308324265397836e-06,
      "loss": 3.6978,
      "step": 220
    },
    {
      "epoch": 0.031953320366768546,
      "grad_norm": 1.2041010856628418,
      "learning_rate": 3.270513696097055e-06,
      "loss": 3.7233,
      "step": 230
    },
    {
      "epoch": 0.0333425951653237,
      "grad_norm": 3.732804536819458,
      "learning_rate": 8.209986176753948e-07,
      "loss": 3.6758,
      "step": 240
    },
    {
      "epoch": 0.03473186996387886,
      "grad_norm": 1.107755422592163,
      "learning_rate": 0.0,
      "loss": 3.7334,
      "step": 250
    }
  ],
  "logging_steps": 10,
  "max_steps": 250,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 63,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.5182045544185856e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}