last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.15,
  "eval_steps": 10,
  "global_step": 39,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0038461538461538464,
      "eval_loss": 3.8364155292510986,
      "eval_runtime": 4.3367,
      "eval_samples_per_second": 25.365,
      "eval_steps_per_second": 12.683,
      "step": 1
    },
    {
      "epoch": 0.019230769230769232,
      "grad_norm": 15.092826843261719,
      "learning_rate": 5e-05,
      "loss": 3.7393,
      "step": 5
    },
    {
      "epoch": 0.038461538461538464,
      "grad_norm": 6.490390300750732,
      "learning_rate": 0.0001,
      "loss": 3.1124,
      "step": 10
    },
    {
      "epoch": 0.038461538461538464,
      "eval_loss": 2.4712672233581543,
      "eval_runtime": 3.609,
      "eval_samples_per_second": 30.479,
      "eval_steps_per_second": 15.24,
      "step": 10
    },
    {
      "epoch": 0.057692307692307696,
      "grad_norm": 4.392829418182373,
      "learning_rate": 9.619397662556435e-05,
      "loss": 2.4486,
      "step": 15
    },
    {
      "epoch": 0.07692307692307693,
      "grad_norm": 2.927060842514038,
      "learning_rate": 8.535533905932738e-05,
      "loss": 1.8888,
      "step": 20
    },
    {
      "epoch": 0.07692307692307693,
      "eval_loss": 1.6859116554260254,
      "eval_runtime": 3.7107,
      "eval_samples_per_second": 29.644,
      "eval_steps_per_second": 14.822,
      "step": 20
    },
    {
      "epoch": 0.09615384615384616,
      "grad_norm": 3.1301217079162598,
      "learning_rate": 6.91341716182545e-05,
      "loss": 1.9827,
      "step": 25
    },
    {
      "epoch": 0.11538461538461539,
      "grad_norm": 2.6760478019714355,
      "learning_rate": 5e-05,
      "loss": 1.5476,
      "step": 30
    },
    {
      "epoch": 0.11538461538461539,
      "eval_loss": 1.5842074155807495,
      "eval_runtime": 3.8685,
      "eval_samples_per_second": 28.434,
      "eval_steps_per_second": 14.217,
      "step": 30
    },
    {
      "epoch": 0.1346153846153846,
      "grad_norm": 3.7138242721557617,
      "learning_rate": 3.086582838174551e-05,
      "loss": 1.6332,
      "step": 35
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 347248838836224.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}