{
  "best_metric": 1.06145525,
  "best_model_checkpoint": "/data/coding/ms-swift/output/v7-20250220-132503/checkpoint-108",
  "epoch": 1.9829351535836177,
  "eval_steps": 50,
  "global_step": 108,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01820250284414107,
      "grad_norm": 6.1875,
      "learning_rate": 1.6666666666666667e-06,
      "loss": 1.5638728141784668,
      "memory(GiB)": 29.73,
      "step": 1,
      "token_acc": 0.6939769707705934,
      "train_speed(iter/s)": 0.129066
    },
    {
      "epoch": 0.09101251422070535,
      "grad_norm": 4.96875,
      "learning_rate": 8.333333333333334e-06,
      "loss": 1.518846035003662,
      "memory(GiB)": 41.66,
      "step": 5,
      "token_acc": 0.7012233049968899,
      "train_speed(iter/s)": 0.226379
    },
    {
      "epoch": 0.1820250284414107,
      "grad_norm": 3.515625,
      "learning_rate": 9.96210254835968e-06,
      "loss": 1.3327471733093261,
      "memory(GiB)": 41.66,
      "step": 10,
      "token_acc": 0.7161527878935017,
      "train_speed(iter/s)": 0.240583
    },
    {
      "epoch": 0.27303754266211605,
      "grad_norm": 2.296875,
      "learning_rate": 9.809128215864096e-06,
      "loss": 1.1619236946105957,
      "memory(GiB)": 45.97,
      "step": 15,
      "token_acc": 0.7329187688216199,
      "train_speed(iter/s)": 0.245364
    },
    {
      "epoch": 0.3640500568828214,
      "grad_norm": 2.296875,
      "learning_rate": 9.542326359097619e-06,
      "loss": 1.1952880859375,
      "memory(GiB)": 45.97,
      "step": 20,
      "token_acc": 0.7218057637847742,
      "train_speed(iter/s)": 0.252196
    },
    {
      "epoch": 0.4550625711035267,
      "grad_norm": 2.328125,
      "learning_rate": 9.168011926105598e-06,
      "loss": 1.111426544189453,
      "memory(GiB)": 45.97,
      "step": 25,
      "token_acc": 0.7425333872925941,
      "train_speed(iter/s)": 0.251882
    },
    {
      "epoch": 0.5460750853242321,
      "grad_norm": 2.578125,
      "learning_rate": 8.695044586103297e-06,
      "loss": 1.1071309089660644,
      "memory(GiB)": 45.97,
      "step": 30,
      "token_acc": 0.7414093361083974,
      "train_speed(iter/s)": 0.254082
    },
    {
      "epoch": 0.6370875995449374,
      "grad_norm": 2.078125,
      "learning_rate": 8.134619029470535e-06,
      "loss": 1.0020055770874023,
      "memory(GiB)": 45.97,
      "step": 35,
      "token_acc": 0.7621102932675633,
      "train_speed(iter/s)": 0.251894
    },
    {
      "epoch": 0.7281001137656428,
      "grad_norm": 2.359375,
      "learning_rate": 7.500000000000001e-06,
      "loss": 1.0203259468078614,
      "memory(GiB)": 50.29,
      "step": 40,
      "token_acc": 0.7580885395117914,
      "train_speed(iter/s)": 0.252412
    },
    {
      "epoch": 0.8191126279863481,
      "grad_norm": 2.21875,
      "learning_rate": 6.806208330935766e-06,
      "loss": 0.9908183097839356,
      "memory(GiB)": 50.29,
      "step": 45,
      "token_acc": 0.7667093258473352,
      "train_speed(iter/s)": 0.252266
    },
    {
      "epoch": 0.9101251422070534,
      "grad_norm": 2.5,
      "learning_rate": 6.0696654160324875e-06,
      "loss": 1.045759677886963,
      "memory(GiB)": 50.29,
      "step": 50,
      "token_acc": 0.7470934799685781,
      "train_speed(iter/s)": 0.251372
    },
    {
      "epoch": 0.9101251422070534,
      "eval_loss": 1.073840618133545,
      "eval_runtime": 0.6301,
      "eval_samples_per_second": 68.243,
      "eval_steps_per_second": 14.283,
      "eval_token_acc": 0.7637732857709076,
      "step": 50
    },
    {
      "epoch": 1.018202502844141,
      "grad_norm": 3.59375,
      "learning_rate": 5.3078045306697154e-06,
      "loss": 1.223165225982666,
      "memory(GiB)": 57.52,
      "step": 55,
      "token_acc": 0.7527071344595012,
      "train_speed(iter/s)": 0.246897
    },
    {
      "epoch": 1.1092150170648465,
      "grad_norm": 2.328125,
      "learning_rate": 4.53865820268349e-06,
      "loss": 1.0227657318115235,
      "memory(GiB)": 57.52,
      "step": 60,
      "token_acc": 0.755286734276229,
      "train_speed(iter/s)": 0.248832
    },
    {
      "epoch": 1.2002275312855517,
      "grad_norm": 2.03125,
      "learning_rate": 3.7804313994581143e-06,
      "loss": 0.9702803611755371,
      "memory(GiB)": 57.52,
      "step": 65,
      "token_acc": 0.7658987281017519,
      "train_speed(iter/s)": 0.248641
    },
    {
      "epoch": 1.2912400455062572,
      "grad_norm": 2.3125,
      "learning_rate": 3.0510706335366034e-06,
      "loss": 1.0110454559326172,
      "memory(GiB)": 57.52,
      "step": 70,
      "token_acc": 0.7643304928863696,
      "train_speed(iter/s)": 0.248241
    },
    {
      "epoch": 1.3822525597269624,
      "grad_norm": 2.390625,
      "learning_rate": 2.3678391856132203e-06,
      "loss": 0.9339286804199218,
      "memory(GiB)": 57.52,
      "step": 75,
      "token_acc": 0.7785570747468379,
      "train_speed(iter/s)": 0.247963
    },
    {
      "epoch": 1.4732650739476678,
      "grad_norm": 1.953125,
      "learning_rate": 1.746908498978791e-06,
      "loss": 0.9071330070495606,
      "memory(GiB)": 57.52,
      "step": 80,
      "token_acc": 0.7775242441528808,
      "train_speed(iter/s)": 0.248416
    },
    {
      "epoch": 1.5642775881683733,
      "grad_norm": 2.203125,
      "learning_rate": 1.202975416726464e-06,
      "loss": 1.0261162757873534,
      "memory(GiB)": 57.52,
      "step": 85,
      "token_acc": 0.7509206426287888,
      "train_speed(iter/s)": 0.248819
    },
    {
      "epoch": 1.6552901023890785,
      "grad_norm": 2.078125,
      "learning_rate": 7.489143213519301e-07,
      "loss": 0.995113468170166,
      "memory(GiB)": 57.52,
      "step": 90,
      "token_acc": 0.7631283572516636,
      "train_speed(iter/s)": 0.24959
    },
    {
      "epoch": 1.7463026166097837,
      "grad_norm": 2.1875,
      "learning_rate": 3.9547241027523164e-07,
      "loss": 0.9445444107055664,
      "memory(GiB)": 57.52,
      "step": 95,
      "token_acc": 0.7700910688608404,
      "train_speed(iter/s)": 0.249979
    },
    {
      "epoch": 1.8373151308304891,
      "grad_norm": 2.03125,
      "learning_rate": 1.510153198249531e-07,
      "loss": 0.9724701881408692,
      "memory(GiB)": 57.52,
      "step": 100,
      "token_acc": 0.7649354027573051,
      "train_speed(iter/s)": 0.250388
    },
    {
      "epoch": 1.8373151308304891,
      "eval_loss": 1.0617759227752686,
      "eval_runtime": 0.6342,
      "eval_samples_per_second": 67.805,
      "eval_steps_per_second": 14.192,
      "eval_token_acc": 0.7669441141498217,
      "step": 100
    },
    {
      "epoch": 1.9283276450511946,
      "grad_norm": 2.1875,
      "learning_rate": 2.1329118524827662e-08,
      "loss": 0.9729574203491211,
      "memory(GiB)": 57.52,
      "step": 105,
      "token_acc": 0.7660818713450293,
      "train_speed(iter/s)": 0.250386
    },
    {
      "epoch": 1.9829351535836177,
      "eval_loss": 1.061455249786377,
      "eval_runtime": 0.6272,
      "eval_samples_per_second": 68.563,
      "eval_steps_per_second": 14.35,
      "eval_token_acc": 0.7653586999603647,
      "step": 108
    }
  ],
  "logging_steps": 5,
  "max_steps": 108,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 5000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.19180283271168e+16,
  "train_batch_size": 5,
  "trial_name": null,
  "trial_params": null
}