last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.010961907371882707,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0010961907371882709,
      "grad_norm": 0.7946075201034546,
      "learning_rate": 2e-05,
      "loss": 1.3176,
      "step": 1
    },
    {
      "epoch": 0.0010961907371882709,
      "eval_loss": 0.9684665203094482,
      "eval_runtime": 17.5952,
      "eval_samples_per_second": 21.881,
      "eval_steps_per_second": 10.969,
      "step": 1
    },
    {
      "epoch": 0.0021923814743765417,
      "grad_norm": 1.1743450164794922,
      "learning_rate": 4e-05,
      "loss": 0.8162,
      "step": 2
    },
    {
      "epoch": 0.003288572211564812,
      "grad_norm": 0.6633036732673645,
      "learning_rate": 6e-05,
      "loss": 0.5918,
      "step": 3
    },
    {
      "epoch": 0.003288572211564812,
      "eval_loss": 0.9594090580940247,
      "eval_runtime": 15.9817,
      "eval_samples_per_second": 24.09,
      "eval_steps_per_second": 12.076,
      "step": 3
    },
    {
      "epoch": 0.004384762948753083,
      "grad_norm": 1.033018708229065,
      "learning_rate": 8e-05,
      "loss": 0.6168,
      "step": 4
    },
    {
      "epoch": 0.005480953685941353,
      "grad_norm": 1.042454481124878,
      "learning_rate": 0.0001,
      "loss": 0.7466,
      "step": 5
    },
    {
      "epoch": 0.006577144423129624,
      "grad_norm": 0.7746133208274841,
      "learning_rate": 0.00012,
      "loss": 0.4468,
      "step": 6
    },
    {
      "epoch": 0.006577144423129624,
      "eval_loss": 0.7657381892204285,
      "eval_runtime": 16.0029,
      "eval_samples_per_second": 24.058,
      "eval_steps_per_second": 12.06,
      "step": 6
    },
    {
      "epoch": 0.007673335160317895,
      "grad_norm": 1.8144712448120117,
      "learning_rate": 0.00014,
      "loss": 0.6918,
      "step": 7
    },
    {
      "epoch": 0.008769525897506167,
      "grad_norm": 1.2779066562652588,
      "learning_rate": 0.00016,
      "loss": 0.2668,
      "step": 8
    },
    {
      "epoch": 0.009865716634694436,
      "grad_norm": 0.8901863098144531,
      "learning_rate": 0.00018,
      "loss": 1.0162,
      "step": 9
    },
    {
      "epoch": 0.009865716634694436,
      "eval_loss": 0.3736807703971863,
      "eval_runtime": 16.037,
      "eval_samples_per_second": 24.007,
      "eval_steps_per_second": 12.035,
      "step": 9
    },
    {
      "epoch": 0.010961907371882707,
      "grad_norm": 0.6853063106536865,
      "learning_rate": 0.0002,
      "loss": 0.1295,
      "step": 10
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2589389947404288.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}