llama-factory/saves/Llama3.1-70B-Chinese-Chat/checkpoint-175/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9950248756218906,
  "eval_steps": 35,
  "global_step": 175,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.028429282160625444,
      "grad_norm": 14.865596771240234,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 1.887,
      "step": 5
    },
    {
      "epoch": 0.05685856432125089,
      "grad_norm": 8.584558486938477,
      "learning_rate": 2.857142857142857e-05,
      "loss": 1.5936,
      "step": 10
    },
    {
      "epoch": 0.08528784648187633,
      "grad_norm": 4.329889297485352,
      "learning_rate": 4.2857142857142856e-05,
      "loss": 0.5385,
      "step": 15
    },
    {
      "epoch": 0.11371712864250177,
      "grad_norm": 1.5808465480804443,
      "learning_rate": 5.714285714285714e-05,
      "loss": 0.4072,
      "step": 20
    },
    {
      "epoch": 0.14214641080312723,
      "grad_norm": 1.9611109495162964,
      "learning_rate": 7.142857142857143e-05,
      "loss": 0.3629,
      "step": 25
    },
    {
      "epoch": 0.17057569296375266,
      "grad_norm": 0.6985721588134766,
      "learning_rate": 8.571428571428571e-05,
      "loss": 0.3385,
      "step": 30
    },
    {
      "epoch": 0.19900497512437812,
      "grad_norm": 0.5211701989173889,
      "learning_rate": 0.0001,
      "loss": 0.3051,
      "step": 35
    },
    {
      "epoch": 0.19900497512437812,
      "eval_loss": 0.25876832008361816,
      "eval_runtime": 1601.7861,
      "eval_samples_per_second": 1.561,
      "eval_steps_per_second": 1.561,
      "step": 35
    },
    {
      "epoch": 0.22743425728500355,
      "grad_norm": 0.7382380962371826,
      "learning_rate": 9.993784606094612e-05,
      "loss": 0.2997,
      "step": 40
    },
    {
      "epoch": 0.255863539445629,
      "grad_norm": 0.7649598121643066,
      "learning_rate": 9.975153876827008e-05,
      "loss": 0.2794,
      "step": 45
    },
    {
      "epoch": 0.28429282160625446,
      "grad_norm": 0.8737666606903076,
      "learning_rate": 9.944154131125642e-05,
      "loss": 0.2777,
      "step": 50
    },
    {
      "epoch": 0.31272210376687987,
      "grad_norm": 18.472232818603516,
      "learning_rate": 9.900862439242719e-05,
      "loss": 0.2866,
      "step": 55
    },
    {
      "epoch": 0.3411513859275053,
      "grad_norm": 0.9156910181045532,
      "learning_rate": 9.84538643114539e-05,
      "loss": 0.2993,
      "step": 60
    },
    {
      "epoch": 0.3695806680881308,
      "grad_norm": 0.5505309700965881,
      "learning_rate": 9.777864028930705e-05,
      "loss": 0.2459,
      "step": 65
    },
    {
      "epoch": 0.39800995024875624,
      "grad_norm": 0.6314955949783325,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.2726,
      "step": 70
    },
    {
      "epoch": 0.39800995024875624,
      "eval_loss": 0.24190224707126617,
      "eval_runtime": 1601.1971,
      "eval_samples_per_second": 1.561,
      "eval_steps_per_second": 1.561,
      "step": 70
    },
    {
      "epoch": 0.42643923240938164,
      "grad_norm": 0.37724295258522034,
      "learning_rate": 9.607381059352038e-05,
      "loss": 0.2722,
      "step": 75
    },
    {
      "epoch": 0.4548685145700071,
      "grad_norm": 0.5261896848678589,
      "learning_rate": 9.504844339512095e-05,
      "loss": 0.2427,
      "step": 80
    },
    {
      "epoch": 0.48329779673063256,
      "grad_norm": 0.5982173681259155,
      "learning_rate": 9.391107866851143e-05,
      "loss": 0.2672,
      "step": 85
    },
    {
      "epoch": 0.511727078891258,
      "grad_norm": 0.5829192399978638,
      "learning_rate": 9.266454408160779e-05,
      "loss": 0.2195,
      "step": 90
    },
    {
      "epoch": 0.5401563610518835,
      "grad_norm": 0.5800922513008118,
      "learning_rate": 9.131193871579975e-05,
      "loss": 0.2454,
      "step": 95
    },
    {
      "epoch": 0.5685856432125089,
      "grad_norm": 0.6000517010688782,
      "learning_rate": 8.985662536114613e-05,
      "loss": 0.3725,
      "step": 100
    },
    {
      "epoch": 0.5970149253731343,
      "grad_norm": 0.5690158605575562,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.2387,
      "step": 105
    },
    {
      "epoch": 0.5970149253731343,
      "eval_loss": 0.23953676223754883,
      "eval_runtime": 1600.7721,
      "eval_samples_per_second": 1.562,
      "eval_steps_per_second": 1.562,
      "step": 105
    },
    {
      "epoch": 0.6254442075337597,
      "grad_norm": 0.38492974638938904,
      "learning_rate": 8.665259359149132e-05,
      "loss": 0.2324,
      "step": 110
    },
    {
      "epoch": 0.6538734896943852,
      "grad_norm": 0.9776778221130371,
      "learning_rate": 8.491184090430364e-05,
      "loss": 0.2409,
      "step": 115
    },
    {
      "epoch": 0.6823027718550106,
      "grad_norm": 0.6985746622085571,
      "learning_rate": 8.308429187984297e-05,
      "loss": 0.2512,
      "step": 120
    },
    {
      "epoch": 0.7107320540156361,
      "grad_norm": 0.38226836919784546,
      "learning_rate": 8.117449009293668e-05,
      "loss": 0.2347,
      "step": 125
    },
    {
      "epoch": 0.7391613361762616,
      "grad_norm": 0.5587658882141113,
      "learning_rate": 7.91871836117395e-05,
      "loss": 0.2357,
      "step": 130
    },
    {
      "epoch": 0.767590618336887,
      "grad_norm": 0.6417293548583984,
      "learning_rate": 7.712731319328798e-05,
      "loss": 0.2607,
      "step": 135
    },
    {
      "epoch": 0.7960199004975125,
      "grad_norm": 0.3534771502017975,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.2508,
      "step": 140
    },
    {
      "epoch": 0.7960199004975125,
      "eval_loss": 0.22614409029483795,
      "eval_runtime": 1601.6762,
      "eval_samples_per_second": 1.561,
      "eval_steps_per_second": 1.561,
      "step": 140
    },
    {
      "epoch": 0.8244491826581379,
      "grad_norm": 0.4052274525165558,
      "learning_rate": 7.281053286765815e-05,
      "loss": 0.2338,
      "step": 145
    },
    {
      "epoch": 0.8528784648187633,
      "grad_norm": 0.544508695602417,
      "learning_rate": 7.056435515653059e-05,
      "loss": 0.2547,
      "step": 150
    },
    {
      "epoch": 0.8813077469793887,
      "grad_norm": 0.4221990406513214,
      "learning_rate": 6.826705121831976e-05,
      "loss": 0.2481,
      "step": 155
    },
    {
      "epoch": 0.9097370291400142,
      "grad_norm": 0.5856378078460693,
      "learning_rate": 6.592433251258423e-05,
      "loss": 0.2431,
      "step": 160
    },
    {
      "epoch": 0.9381663113006397,
      "grad_norm": 0.8348550796508789,
      "learning_rate": 6.354202340715026e-05,
      "loss": 0.228,
      "step": 165
    },
    {
      "epoch": 0.9665955934612651,
      "grad_norm": 0.45988309383392334,
      "learning_rate": 6.112604669781572e-05,
      "loss": 0.2229,
      "step": 170
    },
    {
      "epoch": 0.9950248756218906,
      "grad_norm": 0.3123863637447357,
      "learning_rate": 5.868240888334653e-05,
      "loss": 0.2356,
      "step": 175
    },
    {
      "epoch": 0.9950248756218906,
      "eval_loss": 0.21728534996509552,
      "eval_runtime": 1603.0018,
      "eval_samples_per_second": 1.56,
      "eval_steps_per_second": 1.56,
      "step": 175
    }
  ],
  "logging_steps": 5,
  "max_steps": 350,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 35,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.794457126568985e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}