last-checkpoint/trainer_state.json

{
  "best_metric": 9.236889839172363,
  "best_model_checkpoint": "miner_id_besimray/checkpoint-5",
  "epoch": 0.002037489812550937,
  "eval_steps": 5,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00020374898125509371,
      "grad_norm": 6342301433462784.0,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 9.0297,
      "step": 1
    },
    {
      "epoch": 0.00020374898125509371,
      "eval_loss": 9.31723690032959,
      "eval_runtime": 839.5639,
      "eval_samples_per_second": 3.077,
      "eval_steps_per_second": 0.308,
      "step": 1
    },
    {
      "epoch": 0.00040749796251018743,
      "grad_norm": Infinity,
      "learning_rate": 4.000000000000001e-06,
      "loss": 10.6943,
      "step": 2
    },
    {
      "epoch": 0.0006112469437652812,
      "grad_norm": 1.1465935952866705e+19,
      "learning_rate": 6e-06,
      "loss": 7.7524,
      "step": 3
    },
    {
      "epoch": 0.0008149959250203749,
      "grad_norm": Infinity,
      "learning_rate": 8.000000000000001e-06,
      "loss": 10.3849,
      "step": 4
    },
    {
      "epoch": 0.0010187449062754685,
      "grad_norm": 1.5911073997232538e+18,
      "learning_rate": 1e-05,
      "loss": 9.734,
      "step": 5
    },
    {
      "epoch": 0.0010187449062754685,
      "eval_loss": 9.236889839172363,
      "eval_runtime": 842.6603,
      "eval_samples_per_second": 3.065,
      "eval_steps_per_second": 0.307,
      "step": 5
    },
    {
      "epoch": 0.0012224938875305623,
      "grad_norm": 7954734698725376.0,
      "learning_rate": 1.2e-05,
      "loss": 8.1799,
      "step": 6
    },
    {
      "epoch": 0.0014262428687856561,
      "grad_norm": Infinity,
      "learning_rate": 1.4e-05,
      "loss": 9.2379,
      "step": 7
    },
    {
      "epoch": 0.0016299918500407497,
      "grad_norm": Infinity,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 11.4264,
      "step": 8
    },
    {
      "epoch": 0.0018337408312958435,
      "grad_norm": 9.681396420271145e+17,
      "learning_rate": 1.8e-05,
      "loss": 8.77,
      "step": 9
    },
    {
      "epoch": 0.002037489812550937,
      "grad_norm": 2.2846267641246515e+17,
      "learning_rate": 2e-05,
      "loss": 10.016,
      "step": 10
    },
    {
      "epoch": 0.002037489812550937,
      "eval_loss": 9.262293815612793,
      "eval_runtime": 843.2301,
      "eval_samples_per_second": 3.063,
      "eval_steps_per_second": 0.307,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 5,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 1
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9185331983155200.0,
  "train_batch_size": 10,
  "trial_name": null,
  "trial_params": null
}