checkpoint-6/trainer_state.json

{
  "best_metric": 1.2972379922866821,
  "best_model_checkpoint": "./mistral-7B-Tinybook-epochs-5-lr-0002/checkpoint-6",
  "epoch": 6.0,
  "eval_steps": 1,
  "global_step": 6,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 6.5535804804898685,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 1.4261,
      "step": 1
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.5357962846755981,
      "eval_runtime": 5.1379,
      "eval_samples_per_second": 0.195,
      "eval_steps_per_second": 0.195,
      "step": 1
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.556183599930382,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.4254,
      "step": 2
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.5169316530227661,
      "eval_runtime": 5.1377,
      "eval_samples_per_second": 0.195,
      "eval_steps_per_second": 0.195,
      "step": 2
    },
    {
      "epoch": 3.0,
      "grad_norm": 5.439154362306879,
      "learning_rate": 6e-06,
      "loss": 1.3801,
      "step": 3
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.4326227903366089,
      "eval_runtime": 5.1426,
      "eval_samples_per_second": 0.194,
      "eval_steps_per_second": 0.194,
      "step": 3
    },
    {
      "epoch": 4.0,
      "grad_norm": 10.287730670257169,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.1051,
      "step": 4
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.5058702230453491,
      "eval_runtime": 5.1475,
      "eval_samples_per_second": 0.194,
      "eval_steps_per_second": 0.194,
      "step": 4
    },
    {
      "epoch": 5.0,
      "grad_norm": 102.47095148595771,
      "learning_rate": 1e-05,
      "loss": 0.9669,
      "step": 5
    },
    {
      "epoch": 5.0,
      "eval_loss": 1.3066877126693726,
      "eval_runtime": 5.1369,
      "eval_samples_per_second": 0.195,
      "eval_steps_per_second": 0.195,
      "step": 5
    },
    {
      "epoch": 6.0,
      "grad_norm": 16.035718150173285,
      "learning_rate": 1.2e-05,
      "loss": 0.7256,
      "step": 6
    },
    {
      "epoch": 6.0,
      "eval_loss": 1.2972379922866821,
      "eval_runtime": 5.143,
      "eval_samples_per_second": 0.194,
      "eval_steps_per_second": 0.194,
      "step": 6
    }
  ],
  "logging_steps": 1,
  "max_steps": 6,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 1,
  "total_flos": 1.3420921537481933e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}