checkpoint-4/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.6956521739130435,
  "eval_steps": 2,
  "global_step": 4,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17,
      "grad_norm": 0.4166610439239046,
      "learning_rate": 2e-05,
      "loss": 0.9332,
      "step": 1
    },
    {
      "epoch": 0.17,
      "eval_loss": 1.0365231037139893,
      "eval_runtime": 11.8979,
      "eval_samples_per_second": 1.681,
      "eval_steps_per_second": 0.84,
      "step": 1
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.42192077388393007,
      "learning_rate": 4e-05,
      "loss": 0.9677,
      "step": 2
    },
    {
      "epoch": 0.35,
      "eval_loss": 1.0336682796478271,
      "eval_runtime": 11.9106,
      "eval_samples_per_second": 1.679,
      "eval_steps_per_second": 0.84,
      "step": 2
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.4249677147944204,
      "learning_rate": 6e-05,
      "loss": 0.9551,
      "step": 3
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.4384769363231807,
      "learning_rate": 8e-05,
      "loss": 0.9337,
      "step": 4
    },
    {
      "epoch": 0.7,
      "eval_loss": 0.9899237751960754,
      "eval_runtime": 11.9484,
      "eval_samples_per_second": 1.674,
      "eval_steps_per_second": 0.837,
      "step": 4
    }
  ],
  "logging_steps": 1,
  "max_steps": 5,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 2,
  "total_flos": 8.152661394613862e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}