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": 0.15314830839633942,
      "learning_rate": 2e-05,
      "loss": 0.7763,
      "step": 1
    },
    {
      "epoch": 0.0001625619767536373,
      "eval_loss": 0.8325836062431335,
      "eval_runtime": 143.7545,
      "eval_samples_per_second": 18.024,
      "eval_steps_per_second": 9.015,
      "step": 1
    },
    {
      "epoch": 0.0003251239535072746,
      "grad_norm": 0.19105945527553558,
      "learning_rate": 4e-05,
      "loss": 0.7594,
      "step": 2
    },
    {
      "epoch": 0.000487685930260912,
      "grad_norm": 0.1864684522151947,
      "learning_rate": 6e-05,
      "loss": 0.8487,
      "step": 3
    },
    {
      "epoch": 0.000487685930260912,
      "eval_loss": 0.8321942687034607,
      "eval_runtime": 127.6851,
      "eval_samples_per_second": 20.292,
      "eval_steps_per_second": 10.15,
      "step": 3
    },
    {
      "epoch": 0.0006502479070145493,
      "grad_norm": 0.17509125173091888,
      "learning_rate": 8e-05,
      "loss": 0.7986,
      "step": 4
    },
    {
      "epoch": 0.0008128098837681866,
      "grad_norm": 0.22162194550037384,
      "learning_rate": 0.0001,
      "loss": 0.8673,
      "step": 5
    },
    {
      "epoch": 0.000975371860521824,
      "grad_norm": 0.20332558453083038,
      "learning_rate": 0.00012,
      "loss": 0.8076,
      "step": 6
    },
    {
      "epoch": 0.000975371860521824,
      "eval_loss": 0.826266884803772,
      "eval_runtime": 127.8557,
      "eval_samples_per_second": 20.265,
      "eval_steps_per_second": 10.136,
      "step": 6
    },
    {
      "epoch": 0.0011379338372754612,
      "grad_norm": 0.23090657591819763,
      "learning_rate": 0.00014,
      "loss": 0.8491,
      "step": 7
    },
    {
      "epoch": 0.0013004958140290985,
      "grad_norm": 0.28002697229385376,
      "learning_rate": 0.00016,
      "loss": 0.8703,
      "step": 8
    },
    {
      "epoch": 0.001463057790782736,
      "grad_norm": 0.2905180752277374,
      "learning_rate": 0.00018,
      "loss": 0.7873,
      "step": 9
    },
    {
      "epoch": 0.001463057790782736,
      "eval_loss": 0.7985858917236328,
      "eval_runtime": 127.3819,
      "eval_samples_per_second": 20.34,
      "eval_steps_per_second": 10.174,
      "step": 9
    },
    {
      "epoch": 0.0016256197675363732,
      "grad_norm": 0.3359471261501312,
      "learning_rate": 0.0002,
      "loss": 0.7936,
      "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": 3324514381332480.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}