last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.005669737774627924,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005669737774627923,
      "grad_norm": 2.592542886734009,
      "learning_rate": 2e-05,
      "loss": 0.4669,
      "step": 1
    },
    {
      "epoch": 0.0005669737774627923,
      "eval_loss": 0.9723110198974609,
      "eval_runtime": 77.3052,
      "eval_samples_per_second": 4.812,
      "eval_steps_per_second": 4.812,
      "step": 1
    },
    {
      "epoch": 0.0011339475549255846,
      "grad_norm": 3.569695234298706,
      "learning_rate": 4e-05,
      "loss": 0.883,
      "step": 2
    },
    {
      "epoch": 0.001700921332388377,
      "grad_norm": 3.7738802433013916,
      "learning_rate": 6e-05,
      "loss": 0.9745,
      "step": 3
    },
    {
      "epoch": 0.001700921332388377,
      "eval_loss": 0.7736910581588745,
      "eval_runtime": 77.259,
      "eval_samples_per_second": 4.815,
      "eval_steps_per_second": 4.815,
      "step": 3
    },
    {
      "epoch": 0.002267895109851169,
      "grad_norm": 3.861501932144165,
      "learning_rate": 8e-05,
      "loss": 0.5839,
      "step": 4
    },
    {
      "epoch": 0.002834868887313962,
      "grad_norm": 2.47592830657959,
      "learning_rate": 0.0001,
      "loss": 0.6247,
      "step": 5
    },
    {
      "epoch": 0.003401842664776754,
      "grad_norm": 2.5194220542907715,
      "learning_rate": 0.00012,
      "loss": 0.3239,
      "step": 6
    },
    {
      "epoch": 0.003401842664776754,
      "eval_loss": 0.22851146757602692,
      "eval_runtime": 77.2786,
      "eval_samples_per_second": 4.814,
      "eval_steps_per_second": 4.814,
      "step": 6
    },
    {
      "epoch": 0.003968816442239546,
      "grad_norm": 0.9900692701339722,
      "learning_rate": 0.00014,
      "loss": 0.0811,
      "step": 7
    },
    {
      "epoch": 0.004535790219702338,
      "grad_norm": 3.4848287105560303,
      "learning_rate": 0.00016,
      "loss": 0.284,
      "step": 8
    },
    {
      "epoch": 0.0051027639971651315,
      "grad_norm": 0.6983240246772766,
      "learning_rate": 0.00018,
      "loss": 0.0497,
      "step": 9
    },
    {
      "epoch": 0.0051027639971651315,
      "eval_loss": 0.14964884519577026,
      "eval_runtime": 77.2678,
      "eval_samples_per_second": 4.814,
      "eval_steps_per_second": 4.814,
      "step": 9
    },
    {
      "epoch": 0.005669737774627924,
      "grad_norm": 1.7596927881240845,
      "learning_rate": 0.0002,
      "loss": 0.1116,
      "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": 7418872835604480.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}