{
"best_metric": 6.83237886428833,
"best_model_checkpoint": "./results/models/checkpoint-25230",
"epoch": 6.0,
"eval_steps": 500,
"global_step": 25230,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.11890606420927467,
"grad_norm": 0.4375,
"learning_rate": 0.003990487514863258,
"loss": 6.8648,
"step": 500
},
{
"epoch": 0.23781212841854935,
"grad_norm": 3.78125,
"learning_rate": 0.003980975029726516,
"loss": 6.8372,
"step": 1000
},
{
"epoch": 0.356718192627824,
"grad_norm": 3.21875,
"learning_rate": 0.003971462544589774,
"loss": 6.872,
"step": 1500
},
{
"epoch": 0.4756242568370987,
"grad_norm": 1.8359375,
"learning_rate": 0.0039619500594530324,
"loss": 6.8835,
"step": 2000
},
{
"epoch": 0.5945303210463734,
"grad_norm": 15.25,
"learning_rate": 0.00395243757431629,
"loss": 6.8784,
"step": 2500
},
{
"epoch": 0.713436385255648,
"grad_norm": 2.328125,
"learning_rate": 0.003942925089179548,
"loss": 6.8964,
"step": 3000
},
{
"epoch": 0.8323424494649228,
"grad_norm": 17.0,
"learning_rate": 0.003933412604042806,
"loss": 6.8779,
"step": 3500
},
{
"epoch": 0.9512485136741974,
"grad_norm": 0.98828125,
"learning_rate": 0.003923900118906064,
"loss": 6.8788,
"step": 4000
},
{
"epoch": 1.0,
"eval_loss": 6.88191556930542,
"eval_runtime": 26.2747,
"eval_samples_per_second": 76.119,
"eval_steps_per_second": 0.609,
"step": 4205
},
{
"epoch": 1.070154577883472,
"grad_norm": 5.1875,
"learning_rate": 0.003914387633769323,
"loss": 6.8778,
"step": 4500
},
{
"epoch": 1.1890606420927468,
"grad_norm": 2.296875,
"learning_rate": 0.00390487514863258,
"loss": 6.8945,
"step": 5000
},
{
"epoch": 1.3079667063020213,
"grad_norm": 2.625,
"learning_rate": 0.0038953626634958384,
"loss": 6.8862,
"step": 5500
},
{
"epoch": 1.426872770511296,
"grad_norm": 0.71875,
"learning_rate": 0.0038858501783590963,
"loss": 6.8846,
"step": 6000
},
{
"epoch": 1.5457788347205708,
"grad_norm": 1.3046875,
"learning_rate": 0.0038763376932223546,
"loss": 6.8757,
"step": 6500
},
{
"epoch": 1.6646848989298455,
"grad_norm": 1.0078125,
"learning_rate": 0.0038668252080856125,
"loss": 6.8802,
"step": 7000
},
{
"epoch": 1.7835909631391202,
"grad_norm": 1.5625,
"learning_rate": 0.0038573127229488704,
"loss": 6.8663,
"step": 7500
},
{
"epoch": 1.9024970273483948,
"grad_norm": 12.75,
"learning_rate": 0.0038478002378121287,
"loss": 6.8694,
"step": 8000
},
{
"epoch": 2.0,
"eval_loss": 6.880330562591553,
"eval_runtime": 23.7249,
"eval_samples_per_second": 84.299,
"eval_steps_per_second": 0.674,
"step": 8410
},
{
"epoch": 2.0214030915576693,
"grad_norm": 14.375,
"learning_rate": 0.003838287752675386,
"loss": 6.8764,
"step": 8500
},
{
"epoch": 2.140309155766944,
"grad_norm": 5.09375,
"learning_rate": 0.0038287752675386444,
"loss": 6.8742,
"step": 9000
},
{
"epoch": 2.2592152199762188,
"grad_norm": 17.125,
"learning_rate": 0.0038192627824019027,
"loss": 6.8829,
"step": 9500
},
{
"epoch": 2.3781212841854935,
"grad_norm": 2.171875,
"learning_rate": 0.0038097502972651606,
"loss": 6.8883,
"step": 10000
},
{
"epoch": 2.4970273483947683,
"grad_norm": 1.0,
"learning_rate": 0.0038002378121284185,
"loss": 6.8851,
"step": 10500
},
{
"epoch": 2.6159334126040426,
"grad_norm": 0.84375,
"learning_rate": 0.003790725326991677,
"loss": 6.8849,
"step": 11000
},
{
"epoch": 2.7348394768133177,
"grad_norm": 2.0625,
"learning_rate": 0.0037812128418549347,
"loss": 6.8784,
"step": 11500
},
{
"epoch": 2.853745541022592,
"grad_norm": 5.78125,
"learning_rate": 0.003771700356718193,
"loss": 6.8769,
"step": 12000
},
{
"epoch": 2.972651605231867,
"grad_norm": 5.125,
"learning_rate": 0.0037621878715814504,
"loss": 6.876,
"step": 12500
},
{
"epoch": 3.0,
"eval_loss": 6.865922451019287,
"eval_runtime": 20.8083,
"eval_samples_per_second": 96.115,
"eval_steps_per_second": 0.769,
"step": 12615
},
{
"epoch": 3.0915576694411415,
"grad_norm": 3.8125,
"learning_rate": 0.0037526753864447087,
"loss": 6.8658,
"step": 13000
},
{
"epoch": 3.2104637336504163,
"grad_norm": 6.21875,
"learning_rate": 0.003743162901307967,
"loss": 6.8673,
"step": 13500
},
{
"epoch": 3.329369797859691,
"grad_norm": 3.71875,
"learning_rate": 0.003733650416171225,
"loss": 6.8629,
"step": 14000
},
{
"epoch": 3.4482758620689653,
"grad_norm": 4.5,
"learning_rate": 0.0037241379310344828,
"loss": 6.866,
"step": 14500
},
{
"epoch": 3.56718192627824,
"grad_norm": 2.546875,
"learning_rate": 0.003714625445897741,
"loss": 6.8627,
"step": 15000
},
{
"epoch": 3.686087990487515,
"grad_norm": 7.0625,
"learning_rate": 0.003705112960760999,
"loss": 6.859,
"step": 15500
},
{
"epoch": 3.8049940546967895,
"grad_norm": 42.0,
"learning_rate": 0.003695600475624257,
"loss": 6.852,
"step": 16000
},
{
"epoch": 3.9239001189060643,
"grad_norm": 24.75,
"learning_rate": 0.0036860879904875147,
"loss": 6.8541,
"step": 16500
},
{
"epoch": 4.0,
"eval_loss": 6.84573221206665,
"eval_runtime": 22.0613,
"eval_samples_per_second": 90.656,
"eval_steps_per_second": 0.725,
"step": 16820
},
{
"epoch": 4.042806183115339,
"grad_norm": 19.75,
"learning_rate": 0.003676575505350773,
"loss": 6.85,
"step": 17000
},
{
"epoch": 4.161712247324614,
"grad_norm": 18.0,
"learning_rate": 0.0036670630202140313,
"loss": 6.8548,
"step": 17500
},
{
"epoch": 4.280618311533888,
"grad_norm": 1.7421875,
"learning_rate": 0.0036575505350772888,
"loss": 6.8497,
"step": 18000
},
{
"epoch": 4.399524375743163,
"grad_norm": 11.25,
"learning_rate": 0.003648038049940547,
"loss": 6.8469,
"step": 18500
},
{
"epoch": 4.5184304399524375,
"grad_norm": 181.0,
"learning_rate": 0.003638525564803805,
"loss": 6.8481,
"step": 19000
},
{
"epoch": 4.637336504161713,
"grad_norm": 13.4375,
"learning_rate": 0.0036290130796670633,
"loss": 6.854,
"step": 19500
},
{
"epoch": 4.756242568370987,
"grad_norm": 19.75,
"learning_rate": 0.003619500594530321,
"loss": 6.848,
"step": 20000
},
{
"epoch": 4.875148632580261,
"grad_norm": 3.71875,
"learning_rate": 0.003609988109393579,
"loss": 6.8439,
"step": 20500
},
{
"epoch": 4.9940546967895365,
"grad_norm": 3.78125,
"learning_rate": 0.0036004756242568373,
"loss": 6.8516,
"step": 21000
},
{
"epoch": 5.0,
"eval_loss": 6.845646381378174,
"eval_runtime": 24.3842,
"eval_samples_per_second": 82.02,
"eval_steps_per_second": 0.656,
"step": 21025
},
{
"epoch": 5.112960760998811,
"grad_norm": 43.0,
"learning_rate": 0.003590963139120095,
"loss": 6.8426,
"step": 21500
},
{
"epoch": 5.231866825208086,
"grad_norm": 14.5,
"learning_rate": 0.003581450653983353,
"loss": 6.8388,
"step": 22000
},
{
"epoch": 5.35077288941736,
"grad_norm": 23.25,
"learning_rate": 0.0035719381688466114,
"loss": 6.843,
"step": 22500
},
{
"epoch": 5.469678953626635,
"grad_norm": 3.546875,
"learning_rate": 0.0035624256837098692,
"loss": 6.8391,
"step": 23000
},
{
"epoch": 5.58858501783591,
"grad_norm": 4.03125,
"learning_rate": 0.003552913198573127,
"loss": 6.838,
"step": 23500
},
{
"epoch": 5.707491082045184,
"grad_norm": 2.953125,
"learning_rate": 0.0035434007134363854,
"loss": 6.8373,
"step": 24000
},
{
"epoch": 5.826397146254459,
"grad_norm": 16.25,
"learning_rate": 0.0035338882282996433,
"loss": 6.8372,
"step": 24500
},
{
"epoch": 5.945303210463734,
"grad_norm": 49.0,
"learning_rate": 0.0035243757431629016,
"loss": 6.8363,
"step": 25000
},
{
"epoch": 6.0,
"eval_loss": 6.83237886428833,
"eval_runtime": 24.8165,
"eval_samples_per_second": 80.591,
"eval_steps_per_second": 0.645,
"step": 25230
}
],
"logging_steps": 500,
"max_steps": 210250,
"num_input_tokens_seen": 0,
"num_train_epochs": 50,
"save_steps": 500,
"stateful_callbacks": {
"EarlyStoppingCallback": {
"args": {
"early_stopping_patience": 3,
"early_stopping_threshold": 0.0
},
"attributes": {
"early_stopping_patience_counter": 0
}
},
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 2.337350063694092e+18,
"train_batch_size": 128,
"trial_name": null,
"trial_params": null
}