last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0221606648199446,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00221606648199446,
      "grad_norm": 2.017103672027588,
      "learning_rate": 2e-05,
      "loss": 3.4392,
      "step": 1
    },
    {
      "epoch": 0.00221606648199446,
      "eval_loss": 0.9273197650909424,
      "eval_runtime": 10.2027,
      "eval_samples_per_second": 18.622,
      "eval_steps_per_second": 9.311,
      "step": 1
    },
    {
      "epoch": 0.00443213296398892,
      "grad_norm": 1.4137961864471436,
      "learning_rate": 4e-05,
      "loss": 2.8142,
      "step": 2
    },
    {
      "epoch": 0.006648199445983379,
      "grad_norm": 1.9104070663452148,
      "learning_rate": 6e-05,
      "loss": 4.1668,
      "step": 3
    },
    {
      "epoch": 0.006648199445983379,
      "eval_loss": 0.9240692853927612,
      "eval_runtime": 10.2082,
      "eval_samples_per_second": 18.612,
      "eval_steps_per_second": 9.306,
      "step": 3
    },
    {
      "epoch": 0.00886426592797784,
      "grad_norm": 1.3865467309951782,
      "learning_rate": 8e-05,
      "loss": 3.0463,
      "step": 4
    },
    {
      "epoch": 0.0110803324099723,
      "grad_norm": 2.4066877365112305,
      "learning_rate": 0.0001,
      "loss": 4.6657,
      "step": 5
    },
    {
      "epoch": 0.013296398891966758,
      "grad_norm": 3.50789737701416,
      "learning_rate": 0.00012,
      "loss": 5.0335,
      "step": 6
    },
    {
      "epoch": 0.013296398891966758,
      "eval_loss": 0.85931396484375,
      "eval_runtime": 10.2375,
      "eval_samples_per_second": 18.559,
      "eval_steps_per_second": 9.28,
      "step": 6
    },
    {
      "epoch": 0.01551246537396122,
      "grad_norm": 3.381432294845581,
      "learning_rate": 0.00014,
      "loss": 3.2327,
      "step": 7
    },
    {
      "epoch": 0.01772853185595568,
      "grad_norm": 3.1606369018554688,
      "learning_rate": 0.00016,
      "loss": 2.4294,
      "step": 8
    },
    {
      "epoch": 0.019944598337950138,
      "grad_norm": 3.3392677307128906,
      "learning_rate": 0.00018,
      "loss": 1.9235,
      "step": 9
    },
    {
      "epoch": 0.019944598337950138,
      "eval_loss": 0.42531922459602356,
      "eval_runtime": 10.2258,
      "eval_samples_per_second": 18.581,
      "eval_steps_per_second": 9.29,
      "step": 9
    },
    {
      "epoch": 0.0221606648199446,
      "grad_norm": 5.396928310394287,
      "learning_rate": 0.0002,
      "loss": 2.3041,
      "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": 3166224305356800.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}