trainer_state.json

{
  "best_metric": 0.8162932395935059,
  "best_model_checkpoint": "out/checkpoint-10",
  "epoch": 0.015841584158415842,
  "eval_steps": 1,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 0.9330146908760071,
      "learning_rate": 0.0002,
      "loss": 1.7298,
      "step": 1
    },
    {
      "epoch": 0.0,
      "eval_loss": 1.4902005195617676,
      "eval_runtime": 735.1257,
      "eval_samples_per_second": 0.382,
      "eval_steps_per_second": 0.382,
      "step": 1
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.1639673709869385,
      "learning_rate": 0.0002,
      "loss": 1.4658,
      "step": 2
    },
    {
      "epoch": 0.0,
      "eval_loss": 1.2683874368667603,
      "eval_runtime": 735.139,
      "eval_samples_per_second": 0.382,
      "eval_steps_per_second": 0.382,
      "step": 2
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.126691222190857,
      "learning_rate": 0.0002,
      "loss": 1.1932,
      "step": 3
    },
    {
      "epoch": 0.0,
      "eval_loss": 1.0989333391189575,
      "eval_runtime": 735.2696,
      "eval_samples_per_second": 0.382,
      "eval_steps_per_second": 0.382,
      "step": 3
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.9368861317634583,
      "learning_rate": 0.0002,
      "loss": 1.1229,
      "step": 4
    },
    {
      "epoch": 0.01,
      "eval_loss": 0.9994365572929382,
      "eval_runtime": 735.2205,
      "eval_samples_per_second": 0.382,
      "eval_steps_per_second": 0.382,
      "step": 4
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.5979511141777039,
      "learning_rate": 0.0002,
      "loss": 1.0212,
      "step": 5
    },
    {
      "epoch": 0.01,
      "eval_loss": 0.947055995464325,
      "eval_runtime": 735.2173,
      "eval_samples_per_second": 0.382,
      "eval_steps_per_second": 0.382,
      "step": 5
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.3878212571144104,
      "learning_rate": 0.0002,
      "loss": 0.9752,
      "step": 6
    },
    {
      "epoch": 0.01,
      "eval_loss": 0.9192918539047241,
      "eval_runtime": 735.2312,
      "eval_samples_per_second": 0.382,
      "eval_steps_per_second": 0.382,
      "step": 6
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.3154258728027344,
      "learning_rate": 0.0002,
      "loss": 0.8726,
      "step": 7
    },
    {
      "epoch": 0.01,
      "eval_loss": 0.8976112604141235,
      "eval_runtime": 735.2259,
      "eval_samples_per_second": 0.382,
      "eval_steps_per_second": 0.382,
      "step": 7
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.34492114186286926,
      "learning_rate": 0.0002,
      "loss": 0.8995,
      "step": 8
    },
    {
      "epoch": 0.01,
      "eval_loss": 0.8734176754951477,
      "eval_runtime": 735.259,
      "eval_samples_per_second": 0.382,
      "eval_steps_per_second": 0.382,
      "step": 8
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.4008582830429077,
      "learning_rate": 0.0002,
      "loss": 0.9564,
      "step": 9
    },
    {
      "epoch": 0.01,
      "eval_loss": 0.8452931046485901,
      "eval_runtime": 735.3081,
      "eval_samples_per_second": 0.382,
      "eval_steps_per_second": 0.382,
      "step": 9
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.4416767954826355,
      "learning_rate": 0.0002,
      "loss": 0.817,
      "step": 10
    },
    {
      "epoch": 0.02,
      "eval_loss": 0.8162932395935059,
      "eval_runtime": 735.2796,
      "eval_samples_per_second": 0.382,
      "eval_steps_per_second": 0.382,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1,
  "total_flos": 1628733135912960.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}