last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.006183335909723296,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0006183335909723295,
      "grad_norm": 1.217848539352417,
      "learning_rate": 2e-05,
      "loss": 2.0123,
      "step": 1
    },
    {
      "epoch": 0.0006183335909723295,
      "eval_loss": 2.2555813789367676,
      "eval_runtime": 22.2037,
      "eval_samples_per_second": 30.671,
      "eval_steps_per_second": 15.358,
      "step": 1
    },
    {
      "epoch": 0.001236667181944659,
      "grad_norm": 1.34280526638031,
      "learning_rate": 4e-05,
      "loss": 1.9352,
      "step": 2
    },
    {
      "epoch": 0.0018550007729169888,
      "grad_norm": 1.6996331214904785,
      "learning_rate": 6e-05,
      "loss": 1.7897,
      "step": 3
    },
    {
      "epoch": 0.0018550007729169888,
      "eval_loss": 2.247314691543579,
      "eval_runtime": 20.5603,
      "eval_samples_per_second": 33.122,
      "eval_steps_per_second": 16.585,
      "step": 3
    },
    {
      "epoch": 0.002473334363889318,
      "grad_norm": 1.6258935928344727,
      "learning_rate": 8e-05,
      "loss": 1.9192,
      "step": 4
    },
    {
      "epoch": 0.003091667954861648,
      "grad_norm": 1.3939274549484253,
      "learning_rate": 0.0001,
      "loss": 1.7298,
      "step": 5
    },
    {
      "epoch": 0.0037100015458339775,
      "grad_norm": 1.857796311378479,
      "learning_rate": 0.00012,
      "loss": 2.3341,
      "step": 6
    },
    {
      "epoch": 0.0037100015458339775,
      "eval_loss": 2.116917133331299,
      "eval_runtime": 20.5429,
      "eval_samples_per_second": 33.15,
      "eval_steps_per_second": 16.599,
      "step": 6
    },
    {
      "epoch": 0.004328335136806307,
      "grad_norm": 1.3601452112197876,
      "learning_rate": 0.00014,
      "loss": 1.8658,
      "step": 7
    },
    {
      "epoch": 0.004946668727778636,
      "grad_norm": 1.3681252002716064,
      "learning_rate": 0.00016,
      "loss": 1.548,
      "step": 8
    },
    {
      "epoch": 0.005565002318750966,
      "grad_norm": 2.4520912170410156,
      "learning_rate": 0.00018,
      "loss": 1.9272,
      "step": 9
    },
    {
      "epoch": 0.005565002318750966,
      "eval_loss": 1.8219963312149048,
      "eval_runtime": 20.5466,
      "eval_samples_per_second": 33.144,
      "eval_steps_per_second": 16.596,
      "step": 9
    },
    {
      "epoch": 0.006183335909723296,
      "grad_norm": 1.6402016878128052,
      "learning_rate": 0.0002,
      "loss": 1.5694,
      "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": 1895803354349568.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}