{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.0644433703882713,
"eval_steps": 25,
"global_step": 50,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0012888674077654262,
"grad_norm": 2.766613245010376,
"learning_rate": 5e-05,
"loss": 1.7738,
"step": 1
},
{
"epoch": 0.0012888674077654262,
"eval_loss": 1.8183860778808594,
"eval_runtime": 1.8889,
"eval_samples_per_second": 26.47,
"eval_steps_per_second": 6.882,
"step": 1
},
{
"epoch": 0.0025777348155308524,
"grad_norm": 3.0379271507263184,
"learning_rate": 0.0001,
"loss": 1.7691,
"step": 2
},
{
"epoch": 0.0038666022232962784,
"grad_norm": 2.022273063659668,
"learning_rate": 9.990365154573717e-05,
"loss": 1.6256,
"step": 3
},
{
"epoch": 0.005155469631061705,
"grad_norm": 2.1837451457977295,
"learning_rate": 9.961501876182148e-05,
"loss": 1.4897,
"step": 4
},
{
"epoch": 0.00644433703882713,
"grad_norm": 1.3172669410705566,
"learning_rate": 9.913533761814537e-05,
"loss": 1.3688,
"step": 5
},
{
"epoch": 0.007733204446592557,
"grad_norm": 1.0160404443740845,
"learning_rate": 9.846666218300807e-05,
"loss": 1.3307,
"step": 6
},
{
"epoch": 0.009022071854357983,
"grad_norm": 0.7926989197731018,
"learning_rate": 9.761185582727977e-05,
"loss": 1.3408,
"step": 7
},
{
"epoch": 0.01031093926212341,
"grad_norm": 0.6409326791763306,
"learning_rate": 9.657457896300791e-05,
"loss": 1.3078,
"step": 8
},
{
"epoch": 0.011599806669888834,
"grad_norm": 0.6306732892990112,
"learning_rate": 9.535927336897098e-05,
"loss": 1.2909,
"step": 9
},
{
"epoch": 0.01288867407765426,
"grad_norm": 0.554283618927002,
"learning_rate": 9.397114317029975e-05,
"loss": 1.2566,
"step": 10
},
{
"epoch": 0.014177541485419687,
"grad_norm": 0.5179027915000916,
"learning_rate": 9.241613255361455e-05,
"loss": 1.278,
"step": 11
},
{
"epoch": 0.015466408893185114,
"grad_norm": 0.5059335231781006,
"learning_rate": 9.070090031310558e-05,
"loss": 1.2939,
"step": 12
},
{
"epoch": 0.01675527630095054,
"grad_norm": 0.4588373005390167,
"learning_rate": 8.883279133655399e-05,
"loss": 1.2758,
"step": 13
},
{
"epoch": 0.018044143708715966,
"grad_norm": 0.4461840093135834,
"learning_rate": 8.681980515339464e-05,
"loss": 1.279,
"step": 14
},
{
"epoch": 0.01933301111648139,
"grad_norm": 0.40647274255752563,
"learning_rate": 8.467056167950311e-05,
"loss": 1.2465,
"step": 15
},
{
"epoch": 0.02062187852424682,
"grad_norm": 0.4707936942577362,
"learning_rate": 8.239426430539243e-05,
"loss": 1.2558,
"step": 16
},
{
"epoch": 0.021910745932012244,
"grad_norm": 0.4289848804473877,
"learning_rate": 8.000066048588211e-05,
"loss": 1.2069,
"step": 17
},
{
"epoch": 0.02319961333977767,
"grad_norm": 0.3997207283973694,
"learning_rate": 7.75e-05,
"loss": 1.2114,
"step": 18
},
{
"epoch": 0.024488480747543097,
"grad_norm": 0.4331037104129791,
"learning_rate": 7.490299105985507e-05,
"loss": 1.2551,
"step": 19
},
{
"epoch": 0.02577734815530852,
"grad_norm": 0.4272749125957489,
"learning_rate": 7.222075445642904e-05,
"loss": 1.199,
"step": 20
},
{
"epoch": 0.02706621556307395,
"grad_norm": 0.4014737904071808,
"learning_rate": 6.946477593864228e-05,
"loss": 1.2502,
"step": 21
},
{
"epoch": 0.028355082970839374,
"grad_norm": 0.40306416153907776,
"learning_rate": 6.664685702961344e-05,
"loss": 1.2364,
"step": 22
},
{
"epoch": 0.029643950378604803,
"grad_norm": 0.4330591559410095,
"learning_rate": 6.377906449072578e-05,
"loss": 1.2051,
"step": 23
},
{
"epoch": 0.030932817786370227,
"grad_norm": 0.4139660596847534,
"learning_rate": 6.087367864990233e-05,
"loss": 1.2276,
"step": 24
},
{
"epoch": 0.03222168519413565,
"grad_norm": 0.42047977447509766,
"learning_rate": 5.794314081535644e-05,
"loss": 1.2147,
"step": 25
},
{
"epoch": 0.03222168519413565,
"eval_loss": 1.168257713317871,
"eval_runtime": 1.8869,
"eval_samples_per_second": 26.498,
"eval_steps_per_second": 6.89,
"step": 25
},
{
"epoch": 0.03351055260190108,
"grad_norm": 0.44489747285842896,
"learning_rate": 5.500000000000001e-05,
"loss": 1.2439,
"step": 26
},
{
"epoch": 0.03479942000966651,
"grad_norm": 0.4347832202911377,
"learning_rate": 5.205685918464356e-05,
"loss": 1.2323,
"step": 27
},
{
"epoch": 0.03608828741743193,
"grad_norm": 0.42812854051589966,
"learning_rate": 4.912632135009769e-05,
"loss": 1.2169,
"step": 28
},
{
"epoch": 0.03737715482519736,
"grad_norm": 0.4089065194129944,
"learning_rate": 4.6220935509274235e-05,
"loss": 1.2039,
"step": 29
},
{
"epoch": 0.03866602223296278,
"grad_norm": 0.3960224688053131,
"learning_rate": 4.3353142970386564e-05,
"loss": 1.1992,
"step": 30
},
{
"epoch": 0.03995488964072821,
"grad_norm": 0.4229053258895874,
"learning_rate": 4.053522406135775e-05,
"loss": 1.2278,
"step": 31
},
{
"epoch": 0.04124375704849364,
"grad_norm": 0.47951075434684753,
"learning_rate": 3.777924554357096e-05,
"loss": 1.2345,
"step": 32
},
{
"epoch": 0.04253262445625906,
"grad_norm": 0.4160752296447754,
"learning_rate": 3.509700894014496e-05,
"loss": 1.2111,
"step": 33
},
{
"epoch": 0.04382149186402449,
"grad_norm": 0.41068562865257263,
"learning_rate": 3.250000000000001e-05,
"loss": 1.1921,
"step": 34
},
{
"epoch": 0.04511035927178991,
"grad_norm": 0.3869587779045105,
"learning_rate": 2.9999339514117912e-05,
"loss": 1.1911,
"step": 35
},
{
"epoch": 0.04639922667955534,
"grad_norm": 0.40918055176734924,
"learning_rate": 2.760573569460757e-05,
"loss": 1.2125,
"step": 36
},
{
"epoch": 0.04768809408732077,
"grad_norm": 0.4092864394187927,
"learning_rate": 2.53294383204969e-05,
"loss": 1.2247,
"step": 37
},
{
"epoch": 0.048976961495086194,
"grad_norm": 0.4173303544521332,
"learning_rate": 2.3180194846605367e-05,
"loss": 1.1986,
"step": 38
},
{
"epoch": 0.05026582890285162,
"grad_norm": 0.42150476574897766,
"learning_rate": 2.1167208663446025e-05,
"loss": 1.2046,
"step": 39
},
{
"epoch": 0.05155469631061704,
"grad_norm": 0.43464186787605286,
"learning_rate": 1.9299099686894423e-05,
"loss": 1.2016,
"step": 40
},
{
"epoch": 0.052843563718382475,
"grad_norm": 0.402226060628891,
"learning_rate": 1.758386744638546e-05,
"loss": 1.1816,
"step": 41
},
{
"epoch": 0.0541324311261479,
"grad_norm": 0.3934412896633148,
"learning_rate": 1.602885682970026e-05,
"loss": 1.1789,
"step": 42
},
{
"epoch": 0.055421298533913324,
"grad_norm": 0.3596249520778656,
"learning_rate": 1.464072663102903e-05,
"loss": 1.1911,
"step": 43
},
{
"epoch": 0.05671016594167875,
"grad_norm": 0.35981810092926025,
"learning_rate": 1.3425421036992098e-05,
"loss": 1.1667,
"step": 44
},
{
"epoch": 0.05799903334944417,
"grad_norm": 0.38236004114151,
"learning_rate": 1.2388144172720251e-05,
"loss": 1.2136,
"step": 45
},
{
"epoch": 0.059287900757209605,
"grad_norm": 0.3796616792678833,
"learning_rate": 1.1533337816991932e-05,
"loss": 1.2152,
"step": 46
},
{
"epoch": 0.06057676816497503,
"grad_norm": 0.3959929943084717,
"learning_rate": 1.0864662381854632e-05,
"loss": 1.1892,
"step": 47
},
{
"epoch": 0.061865635572740454,
"grad_norm": 0.40532541275024414,
"learning_rate": 1.0384981238178534e-05,
"loss": 1.1927,
"step": 48
},
{
"epoch": 0.06315450298050589,
"grad_norm": 0.40751609206199646,
"learning_rate": 1.0096348454262845e-05,
"loss": 1.1964,
"step": 49
},
{
"epoch": 0.0644433703882713,
"grad_norm": 0.4269660413265228,
"learning_rate": 1e-05,
"loss": 1.2291,
"step": 50
},
{
"epoch": 0.0644433703882713,
"eval_loss": 1.1453883647918701,
"eval_runtime": 1.8866,
"eval_samples_per_second": 26.503,
"eval_steps_per_second": 6.891,
"step": 50
}
],
"logging_steps": 1,
"max_steps": 50,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 25,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 1.62480894509056e+17,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}