last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.006167129201356768,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0006167129201356768,
      "grad_norm": 0.9303719997406006,
      "learning_rate": 2e-05,
      "loss": 3.595,
      "step": 1
    },
    {
      "epoch": 0.0006167129201356768,
      "eval_loss": 3.5839953422546387,
      "eval_runtime": 28.059,
      "eval_samples_per_second": 24.342,
      "eval_steps_per_second": 12.189,
      "step": 1
    },
    {
      "epoch": 0.0012334258402713536,
      "grad_norm": 0.9219299554824829,
      "learning_rate": 4e-05,
      "loss": 3.8302,
      "step": 2
    },
    {
      "epoch": 0.0018501387604070306,
      "grad_norm": 0.8761361241340637,
      "learning_rate": 6e-05,
      "loss": 3.5784,
      "step": 3
    },
    {
      "epoch": 0.0018501387604070306,
      "eval_loss": 3.56977915763855,
      "eval_runtime": 26.0113,
      "eval_samples_per_second": 26.258,
      "eval_steps_per_second": 13.148,
      "step": 3
    },
    {
      "epoch": 0.002466851680542707,
      "grad_norm": 0.9990836977958679,
      "learning_rate": 8e-05,
      "loss": 3.5951,
      "step": 4
    },
    {
      "epoch": 0.003083564600678384,
      "grad_norm": 0.909419596195221,
      "learning_rate": 0.0001,
      "loss": 3.1928,
      "step": 5
    },
    {
      "epoch": 0.0037002775208140612,
      "grad_norm": 1.3135660886764526,
      "learning_rate": 0.00012,
      "loss": 3.466,
      "step": 6
    },
    {
      "epoch": 0.0037002775208140612,
      "eval_loss": 3.3780205249786377,
      "eval_runtime": 26.1124,
      "eval_samples_per_second": 26.156,
      "eval_steps_per_second": 13.097,
      "step": 6
    },
    {
      "epoch": 0.004316990440949738,
      "grad_norm": 1.0962787866592407,
      "learning_rate": 0.00014,
      "loss": 3.4215,
      "step": 7
    },
    {
      "epoch": 0.004933703361085414,
      "grad_norm": 1.3430908918380737,
      "learning_rate": 0.00016,
      "loss": 3.2145,
      "step": 8
    },
    {
      "epoch": 0.005550416281221091,
      "grad_norm": 0.6255090236663818,
      "learning_rate": 0.00018,
      "loss": 3.114,
      "step": 9
    },
    {
      "epoch": 0.005550416281221091,
      "eval_loss": 3.1323089599609375,
      "eval_runtime": 26.401,
      "eval_samples_per_second": 25.87,
      "eval_steps_per_second": 12.954,
      "step": 9
    },
    {
      "epoch": 0.006167129201356768,
      "grad_norm": 0.793143093585968,
      "learning_rate": 0.0002,
      "loss": 3.3745,
      "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": 2483818032267264.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}