last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00028650216667263544,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.8650216667263546e-05,
      "grad_norm": 0.30766353011131287,
      "learning_rate": 2e-05,
      "loss": 0.4302,
      "step": 1
    },
    {
      "epoch": 2.8650216667263546e-05,
      "eval_loss": 0.44896185398101807,
      "eval_runtime": 698.7018,
      "eval_samples_per_second": 21.035,
      "eval_steps_per_second": 10.518,
      "step": 1
    },
    {
      "epoch": 5.730043333452709e-05,
      "grad_norm": 0.2867901921272278,
      "learning_rate": 4e-05,
      "loss": 0.3545,
      "step": 2
    },
    {
      "epoch": 8.595065000179063e-05,
      "grad_norm": 0.2878236770629883,
      "learning_rate": 6e-05,
      "loss": 0.4487,
      "step": 3
    },
    {
      "epoch": 8.595065000179063e-05,
      "eval_loss": 0.4459349513053894,
      "eval_runtime": 697.637,
      "eval_samples_per_second": 21.067,
      "eval_steps_per_second": 10.534,
      "step": 3
    },
    {
      "epoch": 0.00011460086666905418,
      "grad_norm": 0.3214547038078308,
      "learning_rate": 8e-05,
      "loss": 0.3464,
      "step": 4
    },
    {
      "epoch": 0.00014325108333631772,
      "grad_norm": 0.4778994619846344,
      "learning_rate": 0.0001,
      "loss": 0.3377,
      "step": 5
    },
    {
      "epoch": 0.00017190130000358127,
      "grad_norm": 0.4470870792865753,
      "learning_rate": 0.00012,
      "loss": 0.4031,
      "step": 6
    },
    {
      "epoch": 0.00017190130000358127,
      "eval_loss": 0.4111326336860657,
      "eval_runtime": 697.9828,
      "eval_samples_per_second": 21.056,
      "eval_steps_per_second": 10.529,
      "step": 6
    },
    {
      "epoch": 0.00020055151667084482,
      "grad_norm": 0.6570426821708679,
      "learning_rate": 0.00014,
      "loss": 0.2786,
      "step": 7
    },
    {
      "epoch": 0.00022920173333810837,
      "grad_norm": 0.6415998935699463,
      "learning_rate": 0.00016,
      "loss": 0.3271,
      "step": 8
    },
    {
      "epoch": 0.0002578519500053719,
      "grad_norm": 0.4199046492576599,
      "learning_rate": 0.00018,
      "loss": 0.3007,
      "step": 9
    },
    {
      "epoch": 0.0002578519500053719,
      "eval_loss": 0.2681237459182739,
      "eval_runtime": 697.8688,
      "eval_samples_per_second": 21.06,
      "eval_steps_per_second": 10.531,
      "step": 9
    },
    {
      "epoch": 0.00028650216667263544,
      "grad_norm": 0.592162013053894,
      "learning_rate": 0.0002,
      "loss": 0.2506,
      "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": 3166204172697600.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}