last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.03590664272890485,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.003590664272890485,
      "grad_norm": 3.039785861968994,
      "learning_rate": 2e-05,
      "loss": 3.6641,
      "step": 1
    },
    {
      "epoch": 0.00718132854578097,
      "grad_norm": 3.269423007965088,
      "learning_rate": 4e-05,
      "loss": 3.8954,
      "step": 2
    },
    {
      "epoch": 0.010771992818671455,
      "grad_norm": 3.3091647624969482,
      "learning_rate": 6e-05,
      "loss": 3.2393,
      "step": 3
    },
    {
      "epoch": 0.010771992818671455,
      "eval_loss": 3.640594482421875,
      "eval_runtime": 5.6442,
      "eval_samples_per_second": 20.907,
      "eval_steps_per_second": 10.453,
      "step": 3
    },
    {
      "epoch": 0.01436265709156194,
      "grad_norm": 3.2858850955963135,
      "learning_rate": 8e-05,
      "loss": 3.8674,
      "step": 4
    },
    {
      "epoch": 0.017953321364452424,
      "grad_norm": 3.6083462238311768,
      "learning_rate": 0.0001,
      "loss": 3.4297,
      "step": 5
    },
    {
      "epoch": 0.02154398563734291,
      "grad_norm": 4.41225004196167,
      "learning_rate": 0.00012,
      "loss": 2.9816,
      "step": 6
    },
    {
      "epoch": 0.02154398563734291,
      "eval_loss": 2.8424510955810547,
      "eval_runtime": 5.6309,
      "eval_samples_per_second": 20.956,
      "eval_steps_per_second": 10.478,
      "step": 6
    },
    {
      "epoch": 0.025134649910233394,
      "grad_norm": 3.7052934169769287,
      "learning_rate": 0.00014,
      "loss": 2.7152,
      "step": 7
    },
    {
      "epoch": 0.02872531418312388,
      "grad_norm": 3.1852939128875732,
      "learning_rate": 0.00016,
      "loss": 2.0446,
      "step": 8
    },
    {
      "epoch": 0.03231597845601436,
      "grad_norm": 3.198530435562134,
      "learning_rate": 0.00018,
      "loss": 1.6653,
      "step": 9
    },
    {
      "epoch": 0.03231597845601436,
      "eval_loss": 1.148753046989441,
      "eval_runtime": 5.5446,
      "eval_samples_per_second": 21.282,
      "eval_steps_per_second": 10.641,
      "step": 9
    },
    {
      "epoch": 0.03590664272890485,
      "grad_norm": 3.975350856781006,
      "learning_rate": 0.0002,
      "loss": 1.0576,
      "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": 1647822923366400.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}