{
  "log": [
    {
      "loss": 1.3434071350097656,
      "grad_norm": 0.8922791481018066,
      "learning_rate": 9.6e-05,
      "epoch": 0.11267605633802817,
      "step": 25
    },
    {
      "loss": 0.14198108673095702,
      "grad_norm": 0.271663099527359,
      "learning_rate": 0.000196,
      "epoch": 0.22535211267605634,
      "step": 50
    },
    {
      "loss": 0.03171151399612427,
      "grad_norm": 0.10967417806386948,
      "learning_rate": 0.00019054482374931467,
      "epoch": 0.3380281690140845,
      "step": 75
    },
    {
      "loss": 0.013165457248687744,
      "grad_norm": 0.13412915170192719,
      "learning_rate": 0.00016255277069876454,
      "epoch": 0.4507042253521127,
      "step": 100
    },
    {
      "loss": 0.00924724578857422,
      "grad_norm": 0.22481456398963928,
      "learning_rate": 0.00012174301755815571,
      "epoch": 0.5633802816901409,
      "step": 125
    },
    {
      "loss": 0.008159435987472534,
      "grad_norm": 0.06204904243350029,
      "learning_rate": 7.647789857203645e-05,
      "epoch": 0.676056338028169,
      "step": 150
    },
    {
      "eval_loss": 0.008064636960625648,
      "eval_runtime": 13.0155,
      "eval_samples_per_second": 3.842,
      "eval_steps_per_second": 3.842,
      "epoch": 0.676056338028169,
      "step": 150
    },
    {
      "loss": 0.00772131085395813,
      "grad_norm": 0.06771882623434067,
      "learning_rate": 3.60326978441109e-05,
      "epoch": 0.7887323943661971,
      "step": 175
    },
    {
      "loss": 0.007549132108688354,
      "grad_norm": 0.052400171756744385,
      "learning_rate": 8.695049115005837e-06,
      "epoch": 0.9014084507042254,
      "step": 200
    },
    {
      "eval_loss": 0.007431752514094114,
      "eval_runtime": 13.0193,
      "eval_samples_per_second": 3.84,
      "eval_steps_per_second": 3.84,
      "epoch": 1.0,
      "step": 222
    },
    {
      "train_runtime": 2871.4685,
      "train_samples_per_second": 1.236,
      "train_steps_per_second": 0.077,
      "total_flos": 9372757939200000.0,
      "train_loss": 0.1767391971908174,
      "epoch": 1.0,
      "step": 222
    },
    {
      "eval_loss": 0.007431752514094114,
      "eval_runtime": 12.9964,
      "eval_samples_per_second": 3.847,
      "eval_steps_per_second": 3.847,
      "epoch": 1.0,
      "step": 222
    }
  ],
  "eval": {
    "eval_loss": 0.007431752514094114,
    "eval_runtime": 12.9964,
    "eval_samples_per_second": 3.847,
    "eval_steps_per_second": 3.847,
    "epoch": 1.0
  },
  "base": "Qwen/Qwen2.5-0.5B-Instruct",
  "prev": "Pedro21613/PS-design-1.2-beta",
  "max_len": 1200
}