last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0098015192354815,
  "eval_steps": 1,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00098015192354815,
      "grad_norm": 0.9131542444229126,
      "learning_rate": 2e-05,
      "loss": 1.3184,
      "step": 1
    },
    {
      "epoch": 0.00098015192354815,
      "eval_loss": 1.3775441646575928,
      "eval_runtime": 109.0639,
      "eval_samples_per_second": 3.943,
      "eval_steps_per_second": 1.971,
      "step": 1
    },
    {
      "epoch": 0.0019603038470963,
      "grad_norm": 1.0958034992218018,
      "learning_rate": 4e-05,
      "loss": 1.2625,
      "step": 2
    },
    {
      "epoch": 0.0019603038470963,
      "eval_loss": 1.3744921684265137,
      "eval_runtime": 110.1265,
      "eval_samples_per_second": 3.905,
      "eval_steps_per_second": 1.952,
      "step": 2
    },
    {
      "epoch": 0.00294045577064445,
      "grad_norm": 1.2312837839126587,
      "learning_rate": 6e-05,
      "loss": 1.4153,
      "step": 3
    },
    {
      "epoch": 0.00294045577064445,
      "eval_loss": 1.3623898029327393,
      "eval_runtime": 109.9409,
      "eval_samples_per_second": 3.911,
      "eval_steps_per_second": 1.956,
      "step": 3
    },
    {
      "epoch": 0.0039206076941926,
      "grad_norm": 1.022675633430481,
      "learning_rate": 8e-05,
      "loss": 1.1229,
      "step": 4
    },
    {
      "epoch": 0.0039206076941926,
      "eval_loss": 1.3267920017242432,
      "eval_runtime": 109.9706,
      "eval_samples_per_second": 3.91,
      "eval_steps_per_second": 1.955,
      "step": 4
    },
    {
      "epoch": 0.00490075961774075,
      "grad_norm": 1.2139179706573486,
      "learning_rate": 0.0001,
      "loss": 1.2239,
      "step": 5
    },
    {
      "epoch": 0.00490075961774075,
      "eval_loss": 1.2546281814575195,
      "eval_runtime": 109.9753,
      "eval_samples_per_second": 3.91,
      "eval_steps_per_second": 1.955,
      "step": 5
    },
    {
      "epoch": 0.0058809115412889,
      "grad_norm": 1.0704580545425415,
      "learning_rate": 0.00012,
      "loss": 1.1991,
      "step": 6
    },
    {
      "epoch": 0.0058809115412889,
      "eval_loss": 1.170259714126587,
      "eval_runtime": 109.9581,
      "eval_samples_per_second": 3.911,
      "eval_steps_per_second": 1.955,
      "step": 6
    },
    {
      "epoch": 0.00686106346483705,
      "grad_norm": 0.8198474049568176,
      "learning_rate": 0.00014,
      "loss": 0.9483,
      "step": 7
    },
    {
      "epoch": 0.00686106346483705,
      "eval_loss": 1.0860397815704346,
      "eval_runtime": 109.931,
      "eval_samples_per_second": 3.912,
      "eval_steps_per_second": 1.956,
      "step": 7
    },
    {
      "epoch": 0.0078412153883852,
      "grad_norm": 0.9375729560852051,
      "learning_rate": 0.00016,
      "loss": 1.1308,
      "step": 8
    },
    {
      "epoch": 0.0078412153883852,
      "eval_loss": 0.9986745715141296,
      "eval_runtime": 109.9157,
      "eval_samples_per_second": 3.912,
      "eval_steps_per_second": 1.956,
      "step": 8
    },
    {
      "epoch": 0.008821367311933349,
      "grad_norm": 0.9528597593307495,
      "learning_rate": 0.00018,
      "loss": 1.0389,
      "step": 9
    },
    {
      "epoch": 0.008821367311933349,
      "eval_loss": 0.915632426738739,
      "eval_runtime": 109.9265,
      "eval_samples_per_second": 3.912,
      "eval_steps_per_second": 1.956,
      "step": 9
    },
    {
      "epoch": 0.0098015192354815,
      "grad_norm": 0.9246414303779602,
      "learning_rate": 0.0002,
      "loss": 0.911,
      "step": 10
    },
    {
      "epoch": 0.0098015192354815,
      "eval_loss": 0.8620291948318481,
      "eval_runtime": 109.9074,
      "eval_samples_per_second": 3.912,
      "eval_steps_per_second": 1.956,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7398256992583680.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}