last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.025518341307814992,
  "eval_steps": 5,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0017012227538543328,
      "grad_norm": 0.2222743034362793,
      "learning_rate": 1e-05,
      "loss": 0.7053,
      "step": 1
    },
    {
      "epoch": 0.0017012227538543328,
      "eval_loss": 0.7156262993812561,
      "eval_runtime": 207.1392,
      "eval_samples_per_second": 4.779,
      "eval_steps_per_second": 2.39,
      "step": 1
    },
    {
      "epoch": 0.0034024455077086655,
      "grad_norm": 0.2363133579492569,
      "learning_rate": 2e-05,
      "loss": 0.7089,
      "step": 2
    },
    {
      "epoch": 0.005103668261562998,
      "grad_norm": 0.2022981345653534,
      "learning_rate": 3e-05,
      "loss": 0.671,
      "step": 3
    },
    {
      "epoch": 0.006804891015417331,
      "grad_norm": 0.22789953649044037,
      "learning_rate": 4e-05,
      "loss": 0.6883,
      "step": 4
    },
    {
      "epoch": 0.008506113769271665,
      "grad_norm": 0.21427476406097412,
      "learning_rate": 5e-05,
      "loss": 0.6774,
      "step": 5
    },
    {
      "epoch": 0.008506113769271665,
      "eval_loss": 0.7012587785720825,
      "eval_runtime": 115.8496,
      "eval_samples_per_second": 8.546,
      "eval_steps_per_second": 4.273,
      "step": 5
    },
    {
      "epoch": 0.010207336523125997,
      "grad_norm": 0.26034078001976013,
      "learning_rate": 6e-05,
      "loss": 0.6896,
      "step": 6
    },
    {
      "epoch": 0.01190855927698033,
      "grad_norm": 0.24086257815361023,
      "learning_rate": 7e-05,
      "loss": 0.6932,
      "step": 7
    },
    {
      "epoch": 0.013609782030834662,
      "grad_norm": 0.26397573947906494,
      "learning_rate": 8e-05,
      "loss": 0.6424,
      "step": 8
    },
    {
      "epoch": 0.015311004784688996,
      "grad_norm": 0.2352258712053299,
      "learning_rate": 9e-05,
      "loss": 0.6817,
      "step": 9
    },
    {
      "epoch": 0.01701222753854333,
      "grad_norm": 0.23862141370773315,
      "learning_rate": 0.0001,
      "loss": 0.6594,
      "step": 10
    },
    {
      "epoch": 0.01701222753854333,
      "eval_loss": 0.6520625948905945,
      "eval_runtime": 118.9382,
      "eval_samples_per_second": 8.324,
      "eval_steps_per_second": 4.162,
      "step": 10
    },
    {
      "epoch": 0.01871345029239766,
      "grad_norm": 0.24725323915481567,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.6413,
      "step": 11
    },
    {
      "epoch": 0.020414673046251993,
      "grad_norm": 0.27775460481643677,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.6196,
      "step": 12
    },
    {
      "epoch": 0.022115895800106325,
      "grad_norm": 0.30574458837509155,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.6392,
      "step": 13
    },
    {
      "epoch": 0.02381711855396066,
      "grad_norm": 0.30146658420562744,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.6196,
      "step": 14
    },
    {
      "epoch": 0.025518341307814992,
      "grad_norm": 0.27085185050964355,
      "learning_rate": 5e-05,
      "loss": 0.6135,
      "step": 15
    },
    {
      "epoch": 0.025518341307814992,
      "eval_loss": 0.6110671162605286,
      "eval_runtime": 118.9415,
      "eval_samples_per_second": 8.323,
      "eval_steps_per_second": 4.162,
      "step": 15
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.826444695044096e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}