last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.2127659574468085,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02127659574468085,
      "grad_norm": 3.739654541015625,
      "learning_rate": 2e-05,
      "loss": 4.3413,
      "step": 1
    },
    {
      "epoch": 0.02127659574468085,
      "eval_loss": 3.7383079528808594,
      "eval_runtime": 0.7916,
      "eval_samples_per_second": 25.265,
      "eval_steps_per_second": 12.632,
      "step": 1
    },
    {
      "epoch": 0.0425531914893617,
      "grad_norm": 2.7427635192871094,
      "learning_rate": 4e-05,
      "loss": 4.0764,
      "step": 2
    },
    {
      "epoch": 0.06382978723404255,
      "grad_norm": 2.1136608123779297,
      "learning_rate": 6e-05,
      "loss": 3.8131,
      "step": 3
    },
    {
      "epoch": 0.06382978723404255,
      "eval_loss": 3.731614589691162,
      "eval_runtime": 0.3012,
      "eval_samples_per_second": 66.397,
      "eval_steps_per_second": 33.199,
      "step": 3
    },
    {
      "epoch": 0.0851063829787234,
      "grad_norm": 2.1976680755615234,
      "learning_rate": 8e-05,
      "loss": 3.7946,
      "step": 4
    },
    {
      "epoch": 0.10638297872340426,
      "grad_norm": 2.464569330215454,
      "learning_rate": 0.0001,
      "loss": 3.6771,
      "step": 5
    },
    {
      "epoch": 0.1276595744680851,
      "grad_norm": 3.11846661567688,
      "learning_rate": 0.00012,
      "loss": 4.2779,
      "step": 6
    },
    {
      "epoch": 0.1276595744680851,
      "eval_loss": 3.676224946975708,
      "eval_runtime": 0.3202,
      "eval_samples_per_second": 62.46,
      "eval_steps_per_second": 31.23,
      "step": 6
    },
    {
      "epoch": 0.14893617021276595,
      "grad_norm": 3.397935628890991,
      "learning_rate": 0.00014,
      "loss": 4.021,
      "step": 7
    },
    {
      "epoch": 0.1702127659574468,
      "grad_norm": 2.54946231842041,
      "learning_rate": 0.00016,
      "loss": 3.4491,
      "step": 8
    },
    {
      "epoch": 0.19148936170212766,
      "grad_norm": 2.3854799270629883,
      "learning_rate": 0.00018,
      "loss": 3.4781,
      "step": 9
    },
    {
      "epoch": 0.19148936170212766,
      "eval_loss": 3.475511074066162,
      "eval_runtime": 0.3117,
      "eval_samples_per_second": 64.163,
      "eval_steps_per_second": 32.082,
      "step": 9
    },
    {
      "epoch": 0.2127659574468085,
      "grad_norm": 3.3925769329071045,
      "learning_rate": 0.0002,
      "loss": 3.7794,
      "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": 240545515438080.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}