last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.009992505620784412,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0009992505620784412,
      "grad_norm": 3.7081708908081055,
      "learning_rate": 2e-05,
      "loss": 3.298,
      "step": 1
    },
    {
      "epoch": 0.0019985011241568824,
      "grad_norm": 4.338234901428223,
      "learning_rate": 4e-05,
      "loss": 2.9857,
      "step": 2
    },
    {
      "epoch": 0.0029977516862353237,
      "grad_norm": 5.19654655456543,
      "learning_rate": 6e-05,
      "loss": 3.7834,
      "step": 3
    },
    {
      "epoch": 0.0029977516862353237,
      "eval_loss": 3.598172664642334,
      "eval_runtime": 17.9053,
      "eval_samples_per_second": 23.568,
      "eval_steps_per_second": 11.784,
      "step": 3
    },
    {
      "epoch": 0.003997002248313765,
      "grad_norm": 4.105445384979248,
      "learning_rate": 8e-05,
      "loss": 3.342,
      "step": 4
    },
    {
      "epoch": 0.004996252810392206,
      "grad_norm": 4.069463729858398,
      "learning_rate": 0.0001,
      "loss": 2.9736,
      "step": 5
    },
    {
      "epoch": 0.005995503372470647,
      "grad_norm": 4.118327617645264,
      "learning_rate": 0.00012,
      "loss": 2.9627,
      "step": 6
    },
    {
      "epoch": 0.005995503372470647,
      "eval_loss": 3.472416639328003,
      "eval_runtime": 20.0324,
      "eval_samples_per_second": 21.066,
      "eval_steps_per_second": 10.533,
      "step": 6
    },
    {
      "epoch": 0.0069947539345490885,
      "grad_norm": 4.796413421630859,
      "learning_rate": 0.00014,
      "loss": 3.345,
      "step": 7
    },
    {
      "epoch": 0.00799400449662753,
      "grad_norm": 4.021609783172607,
      "learning_rate": 0.00016,
      "loss": 2.6707,
      "step": 8
    },
    {
      "epoch": 0.008993255058705971,
      "grad_norm": 5.993770599365234,
      "learning_rate": 0.00018,
      "loss": 3.3867,
      "step": 9
    },
    {
      "epoch": 0.008993255058705971,
      "eval_loss": 3.1506168842315674,
      "eval_runtime": 17.998,
      "eval_samples_per_second": 23.447,
      "eval_steps_per_second": 11.723,
      "step": 9
    },
    {
      "epoch": 0.009992505620784412,
      "grad_norm": 5.634053707122803,
      "learning_rate": 0.0002,
      "loss": 3.6112,
      "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": 89038163804160.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}