checkpoint-3932/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 1000,
  "global_step": 3932,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.000254323499491353,
      "grad_norm": 6.461288928985596,
      "learning_rate": 0.0,
      "loss": 3.7722,
      "step": 1
    },
    {
      "epoch": 0.254323499491353,
      "grad_norm": 5.289657115936279,
      "learning_rate": 1.5244150559511698e-05,
      "loss": 2.8712,
      "step": 1000
    },
    {
      "epoch": 0.254323499491353,
      "eval_cosine_accuracy": 0.9444736838340759,
      "eval_loss": 0.5577611327171326,
      "eval_runtime": 35.3135,
      "eval_samples_per_second": 269.274,
      "eval_steps_per_second": 1.076,
      "step": 1000
    },
    {
      "epoch": 0.508646998982706,
      "grad_norm": 6.285799026489258,
      "learning_rate": 2.9944048830111903e-05,
      "loss": 1.8052,
      "step": 2000
    },
    {
      "epoch": 0.508646998982706,
      "eval_cosine_accuracy": 0.9498369693756104,
      "eval_loss": 0.5062179565429688,
      "eval_runtime": 34.9468,
      "eval_samples_per_second": 272.099,
      "eval_steps_per_second": 1.087,
      "step": 2000
    },
    {
      "epoch": 0.762970498474059,
      "grad_norm": 6.981422424316406,
      "learning_rate": 2.8250254323499494e-05,
      "loss": 1.2908,
      "step": 3000
    },
    {
      "epoch": 0.762970498474059,
      "eval_cosine_accuracy": 0.9575139284133911,
      "eval_loss": 0.4583439528942108,
      "eval_runtime": 34.7917,
      "eval_samples_per_second": 273.312,
      "eval_steps_per_second": 1.092,
      "step": 3000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 19660,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}