last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.4444444444444444,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.044444444444444446,
      "grad_norm": 0.3378487527370453,
      "learning_rate": 2e-05,
      "loss": 11.7663,
      "step": 1
    },
    {
      "epoch": 0.044444444444444446,
      "eval_loss": 11.754323959350586,
      "eval_runtime": 0.1112,
      "eval_samples_per_second": 44.98,
      "eval_steps_per_second": 44.98,
      "step": 1
    },
    {
      "epoch": 0.08888888888888889,
      "grad_norm": 0.3135617971420288,
      "learning_rate": 4e-05,
      "loss": 11.7547,
      "step": 2
    },
    {
      "epoch": 0.13333333333333333,
      "grad_norm": 0.4740957021713257,
      "learning_rate": 6e-05,
      "loss": 11.7635,
      "step": 3
    },
    {
      "epoch": 0.13333333333333333,
      "eval_loss": 11.753862380981445,
      "eval_runtime": 0.1092,
      "eval_samples_per_second": 45.802,
      "eval_steps_per_second": 45.802,
      "step": 3
    },
    {
      "epoch": 0.17777777777777778,
      "grad_norm": 0.35033363103866577,
      "learning_rate": 8e-05,
      "loss": 11.7549,
      "step": 4
    },
    {
      "epoch": 0.2222222222222222,
      "grad_norm": 0.3205444812774658,
      "learning_rate": 0.0001,
      "loss": 11.767,
      "step": 5
    },
    {
      "epoch": 0.26666666666666666,
      "grad_norm": 0.3638046085834503,
      "learning_rate": 0.00012,
      "loss": 11.7855,
      "step": 6
    },
    {
      "epoch": 0.26666666666666666,
      "eval_loss": 11.751958847045898,
      "eval_runtime": 0.109,
      "eval_samples_per_second": 45.855,
      "eval_steps_per_second": 45.855,
      "step": 6
    },
    {
      "epoch": 0.3111111111111111,
      "grad_norm": 0.27735668420791626,
      "learning_rate": 0.00014,
      "loss": 11.7312,
      "step": 7
    },
    {
      "epoch": 0.35555555555555557,
      "grad_norm": 0.38196828961372375,
      "learning_rate": 0.00016,
      "loss": 11.7046,
      "step": 8
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.436204731464386,
      "learning_rate": 0.00018,
      "loss": 11.7676,
      "step": 9
    },
    {
      "epoch": 0.4,
      "eval_loss": 11.748748779296875,
      "eval_runtime": 0.1158,
      "eval_samples_per_second": 43.191,
      "eval_steps_per_second": 43.191,
      "step": 9
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 0.3498310446739197,
      "learning_rate": 0.0002,
      "loss": 11.7646,
      "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": 2037000437760.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}