last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.007328691828508612,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0007328691828508611,
      "grad_norm": Infinity,
      "learning_rate": 0.0,
      "loss": 15.4792,
      "step": 1
    },
    {
      "epoch": 0.0007328691828508611,
      "eval_loss": 12.304828643798828,
      "eval_runtime": 57.0786,
      "eval_samples_per_second": 5.046,
      "eval_steps_per_second": 5.046,
      "step": 1
    },
    {
      "epoch": 0.0014657383657017222,
      "grad_norm": 39.946189880371094,
      "learning_rate": 2e-05,
      "loss": 10.6283,
      "step": 2
    },
    {
      "epoch": 0.0021986075485525836,
      "grad_norm": 36.35205078125,
      "learning_rate": 4e-05,
      "loss": 11.3833,
      "step": 3
    },
    {
      "epoch": 0.0021986075485525836,
      "eval_loss": 12.076598167419434,
      "eval_runtime": 56.5009,
      "eval_samples_per_second": 5.097,
      "eval_steps_per_second": 5.097,
      "step": 3
    },
    {
      "epoch": 0.0029314767314034445,
      "grad_norm": 43.041465759277344,
      "learning_rate": 6e-05,
      "loss": 9.7042,
      "step": 4
    },
    {
      "epoch": 0.003664345914254306,
      "grad_norm": 46.06086349487305,
      "learning_rate": 8e-05,
      "loss": 14.8458,
      "step": 5
    },
    {
      "epoch": 0.004397215097105167,
      "grad_norm": 31.989681243896484,
      "learning_rate": 0.0001,
      "loss": 7.5335,
      "step": 6
    },
    {
      "epoch": 0.004397215097105167,
      "eval_loss": 9.594720840454102,
      "eval_runtime": 56.5212,
      "eval_samples_per_second": 5.095,
      "eval_steps_per_second": 5.095,
      "step": 6
    },
    {
      "epoch": 0.005130084279956028,
      "grad_norm": 36.64425277709961,
      "learning_rate": 0.00012,
      "loss": 7.3467,
      "step": 7
    },
    {
      "epoch": 0.005862953462806889,
      "grad_norm": 32.363521575927734,
      "learning_rate": 0.00014,
      "loss": 5.3234,
      "step": 8
    },
    {
      "epoch": 0.00659582264565775,
      "grad_norm": 26.217491149902344,
      "learning_rate": 0.00016,
      "loss": 6.7949,
      "step": 9
    },
    {
      "epoch": 0.00659582264565775,
      "eval_loss": 6.482934951782227,
      "eval_runtime": 56.9284,
      "eval_samples_per_second": 5.059,
      "eval_steps_per_second": 5.059,
      "step": 9
    },
    {
      "epoch": 0.007328691828508612,
      "grad_norm": Infinity,
      "learning_rate": 0.00016,
      "loss": 8.3173,
      "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": 1967550690754560.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}