last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0017287578874578616,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00017287578874578615,
      "grad_norm": 3.553006649017334,
      "learning_rate": 2e-05,
      "loss": 10.147,
      "step": 1
    },
    {
      "epoch": 0.00017287578874578615,
      "eval_loss": 10.197986602783203,
      "eval_runtime": 100.6946,
      "eval_samples_per_second": 24.192,
      "eval_steps_per_second": 12.096,
      "step": 1
    },
    {
      "epoch": 0.0003457515774915723,
      "grad_norm": 3.655125141143799,
      "learning_rate": 4e-05,
      "loss": 9.9026,
      "step": 2
    },
    {
      "epoch": 0.0005186273662373585,
      "grad_norm": 3.4464893341064453,
      "learning_rate": 6e-05,
      "loss": 9.9519,
      "step": 3
    },
    {
      "epoch": 0.0005186273662373585,
      "eval_loss": 10.192020416259766,
      "eval_runtime": 100.4666,
      "eval_samples_per_second": 24.247,
      "eval_steps_per_second": 12.123,
      "step": 3
    },
    {
      "epoch": 0.0006915031549831446,
      "grad_norm": 3.5533785820007324,
      "learning_rate": 8e-05,
      "loss": 10.1142,
      "step": 4
    },
    {
      "epoch": 0.0008643789437289308,
      "grad_norm": 3.568767786026001,
      "learning_rate": 0.0001,
      "loss": 10.1969,
      "step": 5
    },
    {
      "epoch": 0.001037254732474717,
      "grad_norm": 3.5537147521972656,
      "learning_rate": 0.00012,
      "loss": 10.2579,
      "step": 6
    },
    {
      "epoch": 0.001037254732474717,
      "eval_loss": 10.108320236206055,
      "eval_runtime": 105.2129,
      "eval_samples_per_second": 23.153,
      "eval_steps_per_second": 11.577,
      "step": 6
    },
    {
      "epoch": 0.001210130521220503,
      "grad_norm": 3.7763233184814453,
      "learning_rate": 0.00014,
      "loss": 10.4266,
      "step": 7
    },
    {
      "epoch": 0.0013830063099662892,
      "grad_norm": 3.3314616680145264,
      "learning_rate": 0.00016,
      "loss": 10.0965,
      "step": 8
    },
    {
      "epoch": 0.0015558820987120754,
      "grad_norm": 3.2022037506103516,
      "learning_rate": 0.00018,
      "loss": 10.0676,
      "step": 9
    },
    {
      "epoch": 0.0015558820987120754,
      "eval_loss": 9.950599670410156,
      "eval_runtime": 106.7503,
      "eval_samples_per_second": 22.82,
      "eval_steps_per_second": 11.41,
      "step": 9
    },
    {
      "epoch": 0.0017287578874578616,
      "grad_norm": 3.962679147720337,
      "learning_rate": 0.0002,
      "loss": 9.874,
      "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": 5429904015360.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}