last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0002789166875854182,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.7891668758541825e-05,
      "grad_norm": 1.1398605108261108,
      "learning_rate": 2e-05,
      "loss": 4.4271,
      "step": 1
    },
    {
      "epoch": 2.7891668758541825e-05,
      "eval_loss": 1.2013081312179565,
      "eval_runtime": 880.8374,
      "eval_samples_per_second": 17.139,
      "eval_steps_per_second": 8.57,
      "step": 1
    },
    {
      "epoch": 5.578333751708365e-05,
      "grad_norm": 0.9865365028381348,
      "learning_rate": 4e-05,
      "loss": 4.9715,
      "step": 2
    },
    {
      "epoch": 8.367500627562547e-05,
      "grad_norm": 1.1522762775421143,
      "learning_rate": 6e-05,
      "loss": 5.1587,
      "step": 3
    },
    {
      "epoch": 8.367500627562547e-05,
      "eval_loss": 1.1995179653167725,
      "eval_runtime": 880.692,
      "eval_samples_per_second": 17.142,
      "eval_steps_per_second": 8.572,
      "step": 3
    },
    {
      "epoch": 0.0001115666750341673,
      "grad_norm": 1.199426293373108,
      "learning_rate": 8e-05,
      "loss": 5.1546,
      "step": 4
    },
    {
      "epoch": 0.0001394583437927091,
      "grad_norm": 1.3299318552017212,
      "learning_rate": 0.0001,
      "loss": 5.0058,
      "step": 5
    },
    {
      "epoch": 0.00016735001255125095,
      "grad_norm": 1.538432240486145,
      "learning_rate": 0.00012,
      "loss": 5.5428,
      "step": 6
    },
    {
      "epoch": 0.00016735001255125095,
      "eval_loss": 1.166191577911377,
      "eval_runtime": 881.3345,
      "eval_samples_per_second": 17.13,
      "eval_steps_per_second": 8.565,
      "step": 6
    },
    {
      "epoch": 0.00019524168130979276,
      "grad_norm": 2.8678863048553467,
      "learning_rate": 0.00014,
      "loss": 4.2926,
      "step": 7
    },
    {
      "epoch": 0.0002231333500683346,
      "grad_norm": 1.3142216205596924,
      "learning_rate": 0.00016,
      "loss": 4.6445,
      "step": 8
    },
    {
      "epoch": 0.0002510250188268764,
      "grad_norm": 2.5412564277648926,
      "learning_rate": 0.00018,
      "loss": 4.6788,
      "step": 9
    },
    {
      "epoch": 0.0002510250188268764,
      "eval_loss": 1.0516999959945679,
      "eval_runtime": 881.1296,
      "eval_samples_per_second": 17.134,
      "eval_steps_per_second": 8.567,
      "step": 9
    },
    {
      "epoch": 0.0002789166875854182,
      "grad_norm": 1.5285134315490723,
      "learning_rate": 0.0002,
      "loss": 4.2941,
      "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": 3053904828825600.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}