last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.014858841010401188,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0014858841010401188,
      "grad_norm": 23.501811981201172,
      "learning_rate": 2e-05,
      "loss": 10.0714,
      "step": 1
    },
    {
      "epoch": 0.0014858841010401188,
      "eval_loss": 9.684135437011719,
      "eval_runtime": 28.856,
      "eval_samples_per_second": 9.842,
      "eval_steps_per_second": 4.921,
      "step": 1
    },
    {
      "epoch": 0.0029717682020802376,
      "grad_norm": 24.45585823059082,
      "learning_rate": 4e-05,
      "loss": 9.0429,
      "step": 2
    },
    {
      "epoch": 0.004457652303120356,
      "grad_norm": 21.319435119628906,
      "learning_rate": 6e-05,
      "loss": 9.82,
      "step": 3
    },
    {
      "epoch": 0.004457652303120356,
      "eval_loss": 9.280440330505371,
      "eval_runtime": 31.5724,
      "eval_samples_per_second": 8.995,
      "eval_steps_per_second": 4.498,
      "step": 3
    },
    {
      "epoch": 0.005943536404160475,
      "grad_norm": 24.67139434814453,
      "learning_rate": 8e-05,
      "loss": 9.792,
      "step": 4
    },
    {
      "epoch": 0.007429420505200594,
      "grad_norm": 21.461652755737305,
      "learning_rate": 0.0001,
      "loss": 8.7251,
      "step": 5
    },
    {
      "epoch": 0.008915304606240713,
      "grad_norm": 20.924564361572266,
      "learning_rate": 0.00012,
      "loss": 8.7056,
      "step": 6
    },
    {
      "epoch": 0.008915304606240713,
      "eval_loss": 6.362033843994141,
      "eval_runtime": 29.6806,
      "eval_samples_per_second": 9.569,
      "eval_steps_per_second": 4.784,
      "step": 6
    },
    {
      "epoch": 0.010401188707280832,
      "grad_norm": 33.954002380371094,
      "learning_rate": 0.00014,
      "loss": 5.947,
      "step": 7
    },
    {
      "epoch": 0.01188707280832095,
      "grad_norm": 16.948091506958008,
      "learning_rate": 0.00016,
      "loss": 5.0426,
      "step": 8
    },
    {
      "epoch": 0.01337295690936107,
      "grad_norm": 24.004894256591797,
      "learning_rate": 0.00018,
      "loss": 4.6051,
      "step": 9
    },
    {
      "epoch": 0.01337295690936107,
      "eval_loss": 3.463918447494507,
      "eval_runtime": 30.3769,
      "eval_samples_per_second": 9.349,
      "eval_steps_per_second": 4.675,
      "step": 9
    },
    {
      "epoch": 0.014858841010401188,
      "grad_norm": 15.860657691955566,
      "learning_rate": 0.0002,
      "loss": 3.0136,
      "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": 720954300825600.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}