trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 189,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "grad_norm": 5.599682807922363,
      "learning_rate": 0.0001997790438338385,
      "loss": 1.683,
      "num_input_tokens_seen": 4896,
      "step": 5,
      "train_runtime": 118.4256,
      "train_tokens_per_second": 41.342
    },
    {
      "epoch": 0.16,
      "grad_norm": 4.600591659545898,
      "learning_rate": 0.00019888308262251285,
      "loss": 0.9406,
      "num_input_tokens_seen": 10144,
      "step": 10,
      "train_runtime": 247.2151,
      "train_tokens_per_second": 41.033
    },
    {
      "epoch": 0.24,
      "grad_norm": 5.516849994659424,
      "learning_rate": 0.00019730448705798239,
      "loss": 1.1972,
      "num_input_tokens_seen": 14768,
      "step": 15,
      "train_runtime": 367.0408,
      "train_tokens_per_second": 40.235
    },
    {
      "epoch": 0.32,
      "grad_norm": 4.443480968475342,
      "learning_rate": 0.0001950541548947829,
      "loss": 0.8463,
      "num_input_tokens_seen": 19456,
      "step": 20,
      "train_runtime": 488.2663,
      "train_tokens_per_second": 39.847
    },
    {
      "epoch": 0.4,
      "grad_norm": 6.511438846588135,
      "learning_rate": 0.00019214762118704076,
      "loss": 0.9397,
      "num_input_tokens_seen": 24240,
      "step": 25,
      "train_runtime": 611.1324,
      "train_tokens_per_second": 39.664
    },
    {
      "epoch": 0.48,
      "grad_norm": 5.14411735534668,
      "learning_rate": 0.00018860495104301345,
      "loss": 0.8065,
      "num_input_tokens_seen": 28928,
      "step": 30,
      "train_runtime": 732.2813,
      "train_tokens_per_second": 39.504
    },
    {
      "epoch": 0.56,
      "grad_norm": 3.8415708541870117,
      "learning_rate": 0.0001844506011066308,
      "loss": 1.0692,
      "num_input_tokens_seen": 33712,
      "step": 35,
      "train_runtime": 854.4771,
      "train_tokens_per_second": 39.453
    },
    {
      "epoch": 0.64,
      "grad_norm": 4.072882175445557,
      "learning_rate": 0.00017971325072229226,
      "loss": 1.1197,
      "num_input_tokens_seen": 38384,
      "step": 40,
      "train_runtime": 975.508,
      "train_tokens_per_second": 39.348
    },
    {
      "epoch": 0.72,
      "grad_norm": 5.347297191619873,
      "learning_rate": 0.00017442560394846516,
      "loss": 0.9444,
      "num_input_tokens_seen": 43232,
      "step": 45,
      "train_runtime": 1098.6527,
      "train_tokens_per_second": 39.35
    },
    {
      "epoch": 0.8,
      "grad_norm": 4.867551326751709,
      "learning_rate": 0.0001686241637868734,
      "loss": 1.1552,
      "num_input_tokens_seen": 47600,
      "step": 50,
      "train_runtime": 1216.8692,
      "train_tokens_per_second": 39.117
    },
    {
      "epoch": 0.88,
      "grad_norm": 4.36644172668457,
      "learning_rate": 0.00016234898018587337,
      "loss": 0.7799,
      "num_input_tokens_seen": 52416,
      "step": 55,
      "train_runtime": 1339.2743,
      "train_tokens_per_second": 39.138
    },
    {
      "epoch": 0.96,
      "grad_norm": 5.242959499359131,
      "learning_rate": 0.00015564337355766412,
      "loss": 1.0976,
      "num_input_tokens_seen": 57024,
      "step": 60,
      "train_runtime": 1460.6574,
      "train_tokens_per_second": 39.04
    },
    {
      "epoch": 1.032,
      "grad_norm": 2.891428232192993,
      "learning_rate": 0.00014855363571801523,
      "loss": 0.6118,
      "num_input_tokens_seen": 61376,
      "step": 65,
      "train_runtime": 1571.9701,
      "train_tokens_per_second": 39.044
    },
    {
      "epoch": 1.112,
      "grad_norm": 3.863312244415283,
      "learning_rate": 0.00014112871031306119,
      "loss": 0.4582,
      "num_input_tokens_seen": 65808,
      "step": 70,
      "train_runtime": 1688.4702,
      "train_tokens_per_second": 38.975
    },
    {
      "epoch": 1.192,
      "grad_norm": 8.035051345825195,
      "learning_rate": 0.00013341985493931877,
      "loss": 0.4555,
      "num_input_tokens_seen": 70800,
      "step": 75,
      "train_runtime": 1816.0492,
      "train_tokens_per_second": 38.986
    },
    {
      "epoch": 1.272,
      "grad_norm": 6.827272891998291,
      "learning_rate": 0.0001254802872894655,
      "loss": 0.5214,
      "num_input_tokens_seen": 75744,
      "step": 80,
      "train_runtime": 1941.6476,
      "train_tokens_per_second": 39.01
    },
    {
      "epoch": 1.3519999999999999,
      "grad_norm": 3.403914451599121,
      "learning_rate": 0.00011736481776669306,
      "loss": 0.396,
      "num_input_tokens_seen": 80400,
      "step": 85,
      "train_runtime": 2063.1926,
      "train_tokens_per_second": 38.969
    },
    {
      "epoch": 1.432,
      "grad_norm": 5.601797103881836,
      "learning_rate": 0.00010912947110386484,
      "loss": 0.494,
      "num_input_tokens_seen": 85040,
      "step": 90,
      "train_runtime": 2183.6375,
      "train_tokens_per_second": 38.944
    },
    {
      "epoch": 1.512,
      "grad_norm": 4.063144683837891,
      "learning_rate": 0.00010083109959960973,
      "loss": 0.4208,
      "num_input_tokens_seen": 89472,
      "step": 95,
      "train_runtime": 2301.5727,
      "train_tokens_per_second": 38.874
    },
    {
      "epoch": 1.592,
      "grad_norm": 3.933701515197754,
      "learning_rate": 9.252699064135758e-05,
      "loss": 0.3598,
      "num_input_tokens_seen": 94128,
      "step": 100,
      "train_runtime": 2423.5983,
      "train_tokens_per_second": 38.838
    },
    {
      "epoch": 1.6720000000000002,
      "grad_norm": 2.3344547748565674,
      "learning_rate": 8.427447122476148e-05,
      "loss": 0.3386,
      "num_input_tokens_seen": 99120,
      "step": 105,
      "train_runtime": 2556.3976,
      "train_tokens_per_second": 38.773
    },
    {
      "epoch": 1.752,
      "grad_norm": 6.219064712524414,
      "learning_rate": 7.613051219968623e-05,
      "loss": 0.363,
      "num_input_tokens_seen": 104224,
      "step": 110,
      "train_runtime": 2687.8442,
      "train_tokens_per_second": 38.776
    },
    {
      "epoch": 1.8319999999999999,
      "grad_norm": 4.835460662841797,
      "learning_rate": 6.815133497483157e-05,
      "loss": 0.4988,
      "num_input_tokens_seen": 108352,
      "step": 115,
      "train_runtime": 2797.943,
      "train_tokens_per_second": 38.726
    },
    {
      "epoch": 1.912,
      "grad_norm": 3.8654727935791016,
      "learning_rate": 6.039202339608432e-05,
      "loss": 0.492,
      "num_input_tokens_seen": 113216,
      "step": 120,
      "train_runtime": 2921.7812,
      "train_tokens_per_second": 38.749
    },
    {
      "epoch": 1.992,
      "grad_norm": 2.8310158252716064,
      "learning_rate": 5.290614347797802e-05,
      "loss": 0.3521,
      "num_input_tokens_seen": 118128,
      "step": 125,
      "train_runtime": 3048.0813,
      "train_tokens_per_second": 38.755
    },
    {
      "epoch": 2.064,
      "grad_norm": 2.7480006217956543,
      "learning_rate": 4.574537361342407e-05,
      "loss": 0.2833,
      "num_input_tokens_seen": 121984,
      "step": 130,
      "train_runtime": 3150.1518,
      "train_tokens_per_second": 38.723
    },
    {
      "epoch": 2.144,
      "grad_norm": 2.6361961364746094,
      "learning_rate": 3.89591478145437e-05,
      "loss": 0.1752,
      "num_input_tokens_seen": 126272,
      "step": 135,
      "train_runtime": 3263.6021,
      "train_tokens_per_second": 38.691
    },
    {
      "epoch": 2.224,
      "grad_norm": 2.188568115234375,
      "learning_rate": 3.259431444746846e-05,
      "loss": 0.1497,
      "num_input_tokens_seen": 131184,
      "step": 140,
      "train_runtime": 3388.4424,
      "train_tokens_per_second": 38.715
    },
    {
      "epoch": 2.304,
      "grad_norm": 2.502239942550659,
      "learning_rate": 2.669481281701739e-05,
      "loss": 0.123,
      "num_input_tokens_seen": 136256,
      "step": 145,
      "train_runtime": 3517.0188,
      "train_tokens_per_second": 38.742
    },
    {
      "epoch": 2.384,
      "grad_norm": 5.557635307312012,
      "learning_rate": 2.1301369833931117e-05,
      "loss": 0.1846,
      "num_input_tokens_seen": 140816,
      "step": 150,
      "train_runtime": 3636.002,
      "train_tokens_per_second": 38.728
    },
    {
      "epoch": 2.464,
      "grad_norm": 7.517292022705078,
      "learning_rate": 1.6451218858706374e-05,
      "loss": 0.1589,
      "num_input_tokens_seen": 145824,
      "step": 155,
      "train_runtime": 3761.3932,
      "train_tokens_per_second": 38.769
    },
    {
      "epoch": 2.544,
      "grad_norm": 4.320331573486328,
      "learning_rate": 1.2177842662977135e-05,
      "loss": 0.2244,
      "num_input_tokens_seen": 150720,
      "step": 160,
      "train_runtime": 3886.3506,
      "train_tokens_per_second": 38.782
    },
    {
      "epoch": 2.624,
      "grad_norm": 5.886272430419922,
      "learning_rate": 8.510742282896544e-06,
      "loss": 0.243,
      "num_input_tokens_seen": 155408,
      "step": 165,
      "train_runtime": 4010.9688,
      "train_tokens_per_second": 38.746
    },
    {
      "epoch": 2.7039999999999997,
      "grad_norm": 4.565557479858398,
      "learning_rate": 5.475233360227516e-06,
      "loss": 0.1998,
      "num_input_tokens_seen": 159616,
      "step": 170,
      "train_runtime": 4124.7853,
      "train_tokens_per_second": 38.697
    },
    {
      "epoch": 2.784,
      "grad_norm": 5.315474987030029,
      "learning_rate": 3.092271377092215e-06,
      "loss": 0.1511,
      "num_input_tokens_seen": 164544,
      "step": 175,
      "train_runtime": 4249.8174,
      "train_tokens_per_second": 38.718
    },
    {
      "epoch": 2.864,
      "grad_norm": 2.225633382797241,
      "learning_rate": 1.378306990862177e-06,
      "loss": 0.1868,
      "num_input_tokens_seen": 169728,
      "step": 180,
      "train_runtime": 4382.478,
      "train_tokens_per_second": 38.729
    },
    {
      "epoch": 2.944,
      "grad_norm": 3.31820011138916,
      "learning_rate": 3.451724678784518e-07,
      "loss": 0.2023,
      "num_input_tokens_seen": 174320,
      "step": 185,
      "train_runtime": 4501.8922,
      "train_tokens_per_second": 38.721
    },
    {
      "epoch": 3.0,
      "num_input_tokens_seen": 177776,
      "step": 189,
      "total_flos": 7606997652996096.0,
      "train_loss": 0.5502152733071141,
      "train_runtime": 4598.4523,
      "train_samples_per_second": 0.652,
      "train_steps_per_second": 0.041
    }
  ],
  "logging_steps": 5,
  "max_steps": 189,
  "num_input_tokens_seen": 177776,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7606997652996096.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}