last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.010698047606311848,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0010698047606311847,
      "grad_norm": 12.575634002685547,
      "learning_rate": 2e-05,
      "loss": 16.6362,
      "step": 1
    },
    {
      "epoch": 0.0010698047606311847,
      "eval_loss": 17.032079696655273,
      "eval_runtime": 54.0536,
      "eval_samples_per_second": 7.289,
      "eval_steps_per_second": 3.645,
      "step": 1
    },
    {
      "epoch": 0.0021396095212623694,
      "grad_norm": 14.529472351074219,
      "learning_rate": 4e-05,
      "loss": 17.9832,
      "step": 2
    },
    {
      "epoch": 0.0032094142818935543,
      "grad_norm": 12.69963550567627,
      "learning_rate": 6e-05,
      "loss": 17.1144,
      "step": 3
    },
    {
      "epoch": 0.0032094142818935543,
      "eval_loss": 16.841110229492188,
      "eval_runtime": 56.075,
      "eval_samples_per_second": 7.026,
      "eval_steps_per_second": 3.513,
      "step": 3
    },
    {
      "epoch": 0.004279219042524739,
      "grad_norm": 11.762137413024902,
      "learning_rate": 8e-05,
      "loss": 14.5246,
      "step": 4
    },
    {
      "epoch": 0.005349023803155924,
      "grad_norm": 15.911622047424316,
      "learning_rate": 0.0001,
      "loss": 16.0305,
      "step": 5
    },
    {
      "epoch": 0.006418828563787109,
      "grad_norm": 19.69184112548828,
      "learning_rate": 0.00012,
      "loss": 14.2623,
      "step": 6
    },
    {
      "epoch": 0.006418828563787109,
      "eval_loss": 11.754716873168945,
      "eval_runtime": 56.5175,
      "eval_samples_per_second": 6.971,
      "eval_steps_per_second": 3.486,
      "step": 6
    },
    {
      "epoch": 0.007488633324418294,
      "grad_norm": 22.19415283203125,
      "learning_rate": 0.00014,
      "loss": 12.868,
      "step": 7
    },
    {
      "epoch": 0.008558438085049478,
      "grad_norm": 32.429786682128906,
      "learning_rate": 0.00016,
      "loss": 8.5214,
      "step": 8
    },
    {
      "epoch": 0.009628242845680663,
      "grad_norm": 23.76179313659668,
      "learning_rate": 0.00018,
      "loss": 4.399,
      "step": 9
    },
    {
      "epoch": 0.009628242845680663,
      "eval_loss": 2.3507754802703857,
      "eval_runtime": 56.3965,
      "eval_samples_per_second": 6.986,
      "eval_steps_per_second": 3.493,
      "step": 9
    },
    {
      "epoch": 0.010698047606311848,
      "grad_norm": 33.55171203613281,
      "learning_rate": 0.0002,
      "loss": 2.365,
      "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": 2219477097775104.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}