last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0016256197675363732,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0001625619767536373,
      "grad_norm": 2.0558841228485107,
      "learning_rate": 2e-05,
      "loss": 11.9422,
      "step": 1
    },
    {
      "epoch": 0.0001625619767536373,
      "eval_loss": 2.083211660385132,
      "eval_runtime": 70.5571,
      "eval_samples_per_second": 36.722,
      "eval_steps_per_second": 18.368,
      "step": 1
    },
    {
      "epoch": 0.0003251239535072746,
      "grad_norm": 0.6903805732727051,
      "learning_rate": 4e-05,
      "loss": 6.0204,
      "step": 2
    },
    {
      "epoch": 0.000487685930260912,
      "grad_norm": 0.8539021611213684,
      "learning_rate": 6e-05,
      "loss": 7.9875,
      "step": 3
    },
    {
      "epoch": 0.000487685930260912,
      "eval_loss": 2.0831315517425537,
      "eval_runtime": 69.1607,
      "eval_samples_per_second": 37.463,
      "eval_steps_per_second": 18.739,
      "step": 3
    },
    {
      "epoch": 0.0006502479070145493,
      "grad_norm": 1.5361990928649902,
      "learning_rate": 8e-05,
      "loss": 8.7738,
      "step": 4
    },
    {
      "epoch": 0.0008128098837681866,
      "grad_norm": 1.0650941133499146,
      "learning_rate": 0.0001,
      "loss": 7.3005,
      "step": 5
    },
    {
      "epoch": 0.000975371860521824,
      "grad_norm": 1.2282767295837402,
      "learning_rate": 0.00012,
      "loss": 7.6635,
      "step": 6
    },
    {
      "epoch": 0.000975371860521824,
      "eval_loss": 2.0658681392669678,
      "eval_runtime": 69.579,
      "eval_samples_per_second": 37.238,
      "eval_steps_per_second": 18.626,
      "step": 6
    },
    {
      "epoch": 0.0011379338372754612,
      "grad_norm": 2.628141164779663,
      "learning_rate": 0.00014,
      "loss": 8.1998,
      "step": 7
    },
    {
      "epoch": 0.0013004958140290985,
      "grad_norm": 2.1144790649414062,
      "learning_rate": 0.00016,
      "loss": 6.1864,
      "step": 8
    },
    {
      "epoch": 0.001463057790782736,
      "grad_norm": 2.139829635620117,
      "learning_rate": 0.00018,
      "loss": 6.9148,
      "step": 9
    },
    {
      "epoch": 0.001463057790782736,
      "eval_loss": 1.9335415363311768,
      "eval_runtime": 69.3155,
      "eval_samples_per_second": 37.38,
      "eval_steps_per_second": 18.697,
      "step": 9
    },
    {
      "epoch": 0.0016256197675363732,
      "grad_norm": 3.207768440246582,
      "learning_rate": 0.0002,
      "loss": 6.5363,
      "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": 1632478030725120.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}