last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0001293259531322746,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.2932595313227458e-05,
      "grad_norm": 1.1970252990722656,
      "learning_rate": 2e-05,
      "loss": 1.3488,
      "step": 1
    },
    {
      "epoch": 2.5865190626454916e-05,
      "grad_norm": 1.0220004320144653,
      "learning_rate": 4e-05,
      "loss": 1.2877,
      "step": 2
    },
    {
      "epoch": 3.8797785939682375e-05,
      "grad_norm": 1.3500988483428955,
      "learning_rate": 6e-05,
      "loss": 1.5304,
      "step": 3
    },
    {
      "epoch": 3.8797785939682375e-05,
      "eval_loss": 1.4527019262313843,
      "eval_runtime": 1920.9099,
      "eval_samples_per_second": 16.949,
      "eval_steps_per_second": 8.475,
      "step": 3
    },
    {
      "epoch": 5.173038125290983e-05,
      "grad_norm": 1.1525490283966064,
      "learning_rate": 8e-05,
      "loss": 1.4333,
      "step": 4
    },
    {
      "epoch": 6.46629765661373e-05,
      "grad_norm": 1.2471818923950195,
      "learning_rate": 0.0001,
      "loss": 1.2144,
      "step": 5
    },
    {
      "epoch": 7.759557187936475e-05,
      "grad_norm": 1.4231094121932983,
      "learning_rate": 0.00012,
      "loss": 1.3865,
      "step": 6
    },
    {
      "epoch": 7.759557187936475e-05,
      "eval_loss": 1.1237117052078247,
      "eval_runtime": 1909.8201,
      "eval_samples_per_second": 17.048,
      "eval_steps_per_second": 8.524,
      "step": 6
    },
    {
      "epoch": 9.052816719259221e-05,
      "grad_norm": 1.0883755683898926,
      "learning_rate": 0.00014,
      "loss": 0.7752,
      "step": 7
    },
    {
      "epoch": 0.00010346076250581966,
      "grad_norm": 1.611358404159546,
      "learning_rate": 0.00016,
      "loss": 0.8896,
      "step": 8
    },
    {
      "epoch": 0.00011639335781904713,
      "grad_norm": 1.5911290645599365,
      "learning_rate": 0.00018,
      "loss": 0.7958,
      "step": 9
    },
    {
      "epoch": 0.00011639335781904713,
      "eval_loss": 0.48055776953697205,
      "eval_runtime": 1896.4359,
      "eval_samples_per_second": 17.168,
      "eval_steps_per_second": 8.584,
      "step": 9
    },
    {
      "epoch": 0.0001293259531322746,
      "grad_norm": 1.129134178161621,
      "learning_rate": 0.0002,
      "loss": 0.4255,
      "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": 1628734142545920.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}