last-checkpoint/trainer_state.json

{
  "best_metric": 1.4840974807739258,
  "best_model_checkpoint": "miner_id_24/checkpoint-100",
  "epoch": 3.0093457943925235,
  "eval_steps": 100,
  "global_step": 161,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.018691588785046728,
      "eval_loss": 3.1714985370635986,
      "eval_runtime": 1.5135,
      "eval_samples_per_second": 59.464,
      "eval_steps_per_second": 15.196,
      "step": 1
    },
    {
      "epoch": 0.18691588785046728,
      "grad_norm": 4.100635528564453,
      "learning_rate": 0.0002,
      "loss": 2.3359,
      "step": 10
    },
    {
      "epoch": 0.37383177570093457,
      "grad_norm": 1.9233893156051636,
      "learning_rate": 0.00019784350367254322,
      "loss": 1.6888,
      "step": 20
    },
    {
      "epoch": 0.5607476635514018,
      "grad_norm": 1.99177885055542,
      "learning_rate": 0.0001914670242183795,
      "loss": 1.5316,
      "step": 30
    },
    {
      "epoch": 0.7476635514018691,
      "grad_norm": 2.0896918773651123,
      "learning_rate": 0.00018114557872800905,
      "loss": 1.535,
      "step": 40
    },
    {
      "epoch": 0.9345794392523364,
      "grad_norm": 1.9055668115615845,
      "learning_rate": 0.00016732433038731242,
      "loss": 1.4308,
      "step": 50
    },
    {
      "epoch": 1.1214953271028036,
      "grad_norm": 1.394425630569458,
      "learning_rate": 0.00015059938862204127,
      "loss": 1.3753,
      "step": 60
    },
    {
      "epoch": 1.308411214953271,
      "grad_norm": 2.092646598815918,
      "learning_rate": 0.0001316920989420703,
      "loss": 1.1382,
      "step": 70
    },
    {
      "epoch": 1.4953271028037383,
      "grad_norm": 2.206277370452881,
      "learning_rate": 0.00011141793136253986,
      "loss": 1.092,
      "step": 80
    },
    {
      "epoch": 1.6822429906542056,
      "grad_norm": 1.8960262537002563,
      "learning_rate": 9.065130924199998e-05,
      "loss": 1.0111,
      "step": 90
    },
    {
      "epoch": 1.8691588785046729,
      "grad_norm": 1.8522889614105225,
      "learning_rate": 7.028789546718326e-05,
      "loss": 1.0975,
      "step": 100
    },
    {
      "epoch": 1.8691588785046729,
      "eval_loss": 1.4840974807739258,
      "eval_runtime": 1.5931,
      "eval_samples_per_second": 56.494,
      "eval_steps_per_second": 14.437,
      "step": 100
    },
    {
      "epoch": 2.05607476635514,
      "grad_norm": 1.464299201965332,
      "learning_rate": 5.1205962578487155e-05,
      "loss": 1.1104,
      "step": 110
    },
    {
      "epoch": 2.2429906542056073,
      "grad_norm": 2.325424909591675,
      "learning_rate": 3.422851293981676e-05,
      "loss": 0.8096,
      "step": 120
    },
    {
      "epoch": 2.4299065420560746,
      "grad_norm": 1.7495105266571045,
      "learning_rate": 2.008778270707944e-05,
      "loss": 0.7514,
      "step": 130
    },
    {
      "epoch": 2.616822429906542,
      "grad_norm": 1.7600091695785522,
      "learning_rate": 9.393660536564408e-06,
      "loss": 0.7495,
      "step": 140
    },
    {
      "epoch": 2.803738317757009,
      "grad_norm": 1.6766407489776611,
      "learning_rate": 2.607383131993424e-06,
      "loss": 0.7997,
      "step": 150
    },
    {
      "epoch": 2.9906542056074765,
      "grad_norm": 1.9082269668579102,
      "learning_rate": 2.164213936770576e-08,
      "loss": 0.8264,
      "step": 160
    }
  ],
  "logging_steps": 10,
  "max_steps": 161,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 100,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.0229500436545536e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}