7226 lines
177 KiB
JSON
7226 lines
177 KiB
JSON
{
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.9991185843108007,
|
|
"eval_steps": 10,
|
|
"global_step": 921,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"epoch": 0.0010848193097837142,
|
|
"grad_norm": 3.4949746331584786,
|
|
"learning_rate": 3.5714285714285716e-07,
|
|
"loss": 1.1062,
|
|
"step": 1
|
|
},
|
|
{
|
|
"epoch": 0.0021696386195674284,
|
|
"grad_norm": 3.5162997392452056,
|
|
"learning_rate": 7.142857142857143e-07,
|
|
"loss": 1.1094,
|
|
"step": 2
|
|
},
|
|
{
|
|
"epoch": 0.0032544579293511426,
|
|
"grad_norm": 3.5318667768050127,
|
|
"learning_rate": 1.0714285714285714e-06,
|
|
"loss": 1.1104,
|
|
"step": 3
|
|
},
|
|
{
|
|
"epoch": 0.004339277239134857,
|
|
"grad_norm": 3.3609691964965123,
|
|
"learning_rate": 1.4285714285714286e-06,
|
|
"loss": 1.1034,
|
|
"step": 4
|
|
},
|
|
{
|
|
"epoch": 0.005424096548918571,
|
|
"grad_norm": 2.9879203422130147,
|
|
"learning_rate": 1.7857142857142859e-06,
|
|
"loss": 1.1002,
|
|
"step": 5
|
|
},
|
|
{
|
|
"epoch": 0.006508915858702285,
|
|
"grad_norm": 2.8623995866321,
|
|
"learning_rate": 2.1428571428571427e-06,
|
|
"loss": 1.096,
|
|
"step": 6
|
|
},
|
|
{
|
|
"epoch": 0.007593735168485999,
|
|
"grad_norm": 2.3547138531194634,
|
|
"learning_rate": 2.5e-06,
|
|
"loss": 1.0785,
|
|
"step": 7
|
|
},
|
|
{
|
|
"epoch": 0.008678554478269714,
|
|
"grad_norm": 2.2490026217614547,
|
|
"learning_rate": 2.8571428571428573e-06,
|
|
"loss": 1.0725,
|
|
"step": 8
|
|
},
|
|
{
|
|
"epoch": 0.009763373788053427,
|
|
"grad_norm": 2.6803737230607054,
|
|
"learning_rate": 3.2142857142857147e-06,
|
|
"loss": 1.053,
|
|
"step": 9
|
|
},
|
|
{
|
|
"epoch": 0.010848193097837142,
|
|
"grad_norm": 2.4684067036056785,
|
|
"learning_rate": 3.5714285714285718e-06,
|
|
"loss": 1.0465,
|
|
"step": 10
|
|
},
|
|
{
|
|
"epoch": 0.010848193097837142,
|
|
"eval_loss": 1.0118134021759033,
|
|
"eval_runtime": 116.2339,
|
|
"eval_samples_per_second": 82.024,
|
|
"eval_steps_per_second": 0.645,
|
|
"step": 10
|
|
},
|
|
{
|
|
"epoch": 0.011933012407620855,
|
|
"grad_norm": 1.8283207388232985,
|
|
"learning_rate": 3.928571428571429e-06,
|
|
"loss": 1.0131,
|
|
"step": 11
|
|
},
|
|
{
|
|
"epoch": 0.01301783171740457,
|
|
"grad_norm": 1.787912431514352,
|
|
"learning_rate": 4.2857142857142855e-06,
|
|
"loss": 1.0091,
|
|
"step": 12
|
|
},
|
|
{
|
|
"epoch": 0.014102651027188284,
|
|
"grad_norm": 1.4951212244487242,
|
|
"learning_rate": 4.642857142857144e-06,
|
|
"loss": 0.9996,
|
|
"step": 13
|
|
},
|
|
{
|
|
"epoch": 0.015187470336971999,
|
|
"grad_norm": 1.1565998646716686,
|
|
"learning_rate": 5e-06,
|
|
"loss": 0.9873,
|
|
"step": 14
|
|
},
|
|
{
|
|
"epoch": 0.016272289646755714,
|
|
"grad_norm": 1.3885558935074018,
|
|
"learning_rate": 5.357142857142857e-06,
|
|
"loss": 0.9735,
|
|
"step": 15
|
|
},
|
|
{
|
|
"epoch": 0.017357108956539427,
|
|
"grad_norm": 1.496877063571748,
|
|
"learning_rate": 5.7142857142857145e-06,
|
|
"loss": 0.9707,
|
|
"step": 16
|
|
},
|
|
{
|
|
"epoch": 0.01844192826632314,
|
|
"grad_norm": 1.2293672130451796,
|
|
"learning_rate": 6.071428571428571e-06,
|
|
"loss": 0.9619,
|
|
"step": 17
|
|
},
|
|
{
|
|
"epoch": 0.019526747576106854,
|
|
"grad_norm": 1.2444548728652165,
|
|
"learning_rate": 6.4285714285714295e-06,
|
|
"loss": 0.9522,
|
|
"step": 18
|
|
},
|
|
{
|
|
"epoch": 0.02061156688589057,
|
|
"grad_norm": 1.0967934938070116,
|
|
"learning_rate": 6.785714285714287e-06,
|
|
"loss": 0.9446,
|
|
"step": 19
|
|
},
|
|
{
|
|
"epoch": 0.021696386195674284,
|
|
"grad_norm": 1.0868526598858828,
|
|
"learning_rate": 7.1428571428571436e-06,
|
|
"loss": 0.9352,
|
|
"step": 20
|
|
},
|
|
{
|
|
"epoch": 0.021696386195674284,
|
|
"eval_loss": 0.9294480085372925,
|
|
"eval_runtime": 108.6996,
|
|
"eval_samples_per_second": 87.71,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 20
|
|
},
|
|
{
|
|
"epoch": 0.022781205505457997,
|
|
"grad_norm": 0.8412495929358779,
|
|
"learning_rate": 7.500000000000001e-06,
|
|
"loss": 0.9324,
|
|
"step": 21
|
|
},
|
|
{
|
|
"epoch": 0.02386602481524171,
|
|
"grad_norm": 0.9890507743689557,
|
|
"learning_rate": 7.857142857142858e-06,
|
|
"loss": 0.9237,
|
|
"step": 22
|
|
},
|
|
{
|
|
"epoch": 0.024950844125025424,
|
|
"grad_norm": 0.8668353040977067,
|
|
"learning_rate": 8.214285714285714e-06,
|
|
"loss": 0.9182,
|
|
"step": 23
|
|
},
|
|
{
|
|
"epoch": 0.02603566343480914,
|
|
"grad_norm": 0.7309092210427514,
|
|
"learning_rate": 8.571428571428571e-06,
|
|
"loss": 0.9098,
|
|
"step": 24
|
|
},
|
|
{
|
|
"epoch": 0.027120482744592854,
|
|
"grad_norm": 0.7964248954400254,
|
|
"learning_rate": 8.92857142857143e-06,
|
|
"loss": 0.9084,
|
|
"step": 25
|
|
},
|
|
{
|
|
"epoch": 0.028205302054376567,
|
|
"grad_norm": 0.7875083488207993,
|
|
"learning_rate": 9.285714285714288e-06,
|
|
"loss": 0.9014,
|
|
"step": 26
|
|
},
|
|
{
|
|
"epoch": 0.02929012136416028,
|
|
"grad_norm": 0.6361343649039687,
|
|
"learning_rate": 9.642857142857144e-06,
|
|
"loss": 0.8935,
|
|
"step": 27
|
|
},
|
|
{
|
|
"epoch": 0.030374940673943997,
|
|
"grad_norm": 0.7111881277870613,
|
|
"learning_rate": 1e-05,
|
|
"loss": 0.8949,
|
|
"step": 28
|
|
},
|
|
{
|
|
"epoch": 0.03145975998372771,
|
|
"grad_norm": 0.648873533838907,
|
|
"learning_rate": 9.99996905885448e-06,
|
|
"loss": 0.8889,
|
|
"step": 29
|
|
},
|
|
{
|
|
"epoch": 0.03254457929351143,
|
|
"grad_norm": 0.6820782579263859,
|
|
"learning_rate": 9.999876235800859e-06,
|
|
"loss": 0.882,
|
|
"step": 30
|
|
},
|
|
{
|
|
"epoch": 0.03254457929351143,
|
|
"eval_loss": 0.878837525844574,
|
|
"eval_runtime": 108.8059,
|
|
"eval_samples_per_second": 87.624,
|
|
"eval_steps_per_second": 0.689,
|
|
"step": 30
|
|
},
|
|
{
|
|
"epoch": 0.03362939860329514,
|
|
"grad_norm": 0.6419813251816439,
|
|
"learning_rate": 9.999721531987958e-06,
|
|
"loss": 0.8765,
|
|
"step": 31
|
|
},
|
|
{
|
|
"epoch": 0.034714217913078854,
|
|
"grad_norm": 0.611752560846071,
|
|
"learning_rate": 9.99950494933046e-06,
|
|
"loss": 0.8769,
|
|
"step": 32
|
|
},
|
|
{
|
|
"epoch": 0.03579903722286257,
|
|
"grad_norm": 0.551286709663645,
|
|
"learning_rate": 9.999226490508897e-06,
|
|
"loss": 0.8712,
|
|
"step": 33
|
|
},
|
|
{
|
|
"epoch": 0.03688385653264628,
|
|
"grad_norm": 0.5656177125175853,
|
|
"learning_rate": 9.9988861589696e-06,
|
|
"loss": 0.8672,
|
|
"step": 34
|
|
},
|
|
{
|
|
"epoch": 0.037968675842429994,
|
|
"grad_norm": 0.5483165258431487,
|
|
"learning_rate": 9.998483958924666e-06,
|
|
"loss": 0.8677,
|
|
"step": 35
|
|
},
|
|
{
|
|
"epoch": 0.03905349515221371,
|
|
"grad_norm": 0.5453570885370813,
|
|
"learning_rate": 9.99801989535191e-06,
|
|
"loss": 0.86,
|
|
"step": 36
|
|
},
|
|
{
|
|
"epoch": 0.04013831446199742,
|
|
"grad_norm": 0.5723840525108582,
|
|
"learning_rate": 9.997493973994793e-06,
|
|
"loss": 0.8568,
|
|
"step": 37
|
|
},
|
|
{
|
|
"epoch": 0.04122313377178114,
|
|
"grad_norm": 0.592144228561935,
|
|
"learning_rate": 9.996906201362361e-06,
|
|
"loss": 0.8572,
|
|
"step": 38
|
|
},
|
|
{
|
|
"epoch": 0.042307953081564854,
|
|
"grad_norm": 0.6449377321868768,
|
|
"learning_rate": 9.996256584729157e-06,
|
|
"loss": 0.8553,
|
|
"step": 39
|
|
},
|
|
{
|
|
"epoch": 0.04339277239134857,
|
|
"grad_norm": 0.6422868289015168,
|
|
"learning_rate": 9.995545132135133e-06,
|
|
"loss": 0.8491,
|
|
"step": 40
|
|
},
|
|
{
|
|
"epoch": 0.04339277239134857,
|
|
"eval_loss": 0.8473296761512756,
|
|
"eval_runtime": 108.7796,
|
|
"eval_samples_per_second": 87.645,
|
|
"eval_steps_per_second": 0.689,
|
|
"step": 40
|
|
},
|
|
{
|
|
"epoch": 0.04447759170113228,
|
|
"grad_norm": 0.6561423763363042,
|
|
"learning_rate": 9.994771852385552e-06,
|
|
"loss": 0.8516,
|
|
"step": 41
|
|
},
|
|
{
|
|
"epoch": 0.045562411010915994,
|
|
"grad_norm": 0.6891506900391302,
|
|
"learning_rate": 9.993936755050881e-06,
|
|
"loss": 0.8457,
|
|
"step": 42
|
|
},
|
|
{
|
|
"epoch": 0.04664723032069971,
|
|
"grad_norm": 0.5746241844843091,
|
|
"learning_rate": 9.993039850466664e-06,
|
|
"loss": 0.8422,
|
|
"step": 43
|
|
},
|
|
{
|
|
"epoch": 0.04773204963048342,
|
|
"grad_norm": 0.47671534032332025,
|
|
"learning_rate": 9.992081149733404e-06,
|
|
"loss": 0.84,
|
|
"step": 44
|
|
},
|
|
{
|
|
"epoch": 0.048816868940267134,
|
|
"grad_norm": 0.4622095523311291,
|
|
"learning_rate": 9.991060664716423e-06,
|
|
"loss": 0.8355,
|
|
"step": 45
|
|
},
|
|
{
|
|
"epoch": 0.04990168825005085,
|
|
"grad_norm": 0.48695761710710606,
|
|
"learning_rate": 9.989978408045709e-06,
|
|
"loss": 0.8353,
|
|
"step": 46
|
|
},
|
|
{
|
|
"epoch": 0.05098650755983457,
|
|
"grad_norm": 0.5109048866982651,
|
|
"learning_rate": 9.988834393115768e-06,
|
|
"loss": 0.833,
|
|
"step": 47
|
|
},
|
|
{
|
|
"epoch": 0.05207132686961828,
|
|
"grad_norm": 0.4999757467134964,
|
|
"learning_rate": 9.98762863408545e-06,
|
|
"loss": 0.8345,
|
|
"step": 48
|
|
},
|
|
{
|
|
"epoch": 0.053156146179401995,
|
|
"grad_norm": 0.4723225493895438,
|
|
"learning_rate": 9.986361145877783e-06,
|
|
"loss": 0.8283,
|
|
"step": 49
|
|
},
|
|
{
|
|
"epoch": 0.05424096548918571,
|
|
"grad_norm": 0.4212216755382298,
|
|
"learning_rate": 9.985031944179781e-06,
|
|
"loss": 0.8279,
|
|
"step": 50
|
|
},
|
|
{
|
|
"epoch": 0.05424096548918571,
|
|
"eval_loss": 0.8254933953285217,
|
|
"eval_runtime": 108.5477,
|
|
"eval_samples_per_second": 87.832,
|
|
"eval_steps_per_second": 0.691,
|
|
"step": 50
|
|
},
|
|
{
|
|
"epoch": 0.05532578479896942,
|
|
"grad_norm": 0.39527050101497846,
|
|
"learning_rate": 9.983641045442256e-06,
|
|
"loss": 0.8247,
|
|
"step": 51
|
|
},
|
|
{
|
|
"epoch": 0.056410604108753135,
|
|
"grad_norm": 0.3567819565829818,
|
|
"learning_rate": 9.982188466879607e-06,
|
|
"loss": 0.8234,
|
|
"step": 52
|
|
},
|
|
{
|
|
"epoch": 0.05749542341853685,
|
|
"grad_norm": 0.3865472249163068,
|
|
"learning_rate": 9.980674226469608e-06,
|
|
"loss": 0.8219,
|
|
"step": 53
|
|
},
|
|
{
|
|
"epoch": 0.05858024272832056,
|
|
"grad_norm": 0.3675626983233842,
|
|
"learning_rate": 9.979098342953198e-06,
|
|
"loss": 0.8188,
|
|
"step": 54
|
|
},
|
|
{
|
|
"epoch": 0.05966506203810428,
|
|
"grad_norm": 0.38616224402835747,
|
|
"learning_rate": 9.977460835834231e-06,
|
|
"loss": 0.8194,
|
|
"step": 55
|
|
},
|
|
{
|
|
"epoch": 0.060749881347887995,
|
|
"grad_norm": 0.5369962495644119,
|
|
"learning_rate": 9.975761725379243e-06,
|
|
"loss": 0.8194,
|
|
"step": 56
|
|
},
|
|
{
|
|
"epoch": 0.06183470065767171,
|
|
"grad_norm": 0.7610499469721489,
|
|
"learning_rate": 9.974001032617208e-06,
|
|
"loss": 0.8177,
|
|
"step": 57
|
|
},
|
|
{
|
|
"epoch": 0.06291951996745541,
|
|
"grad_norm": 1.1868133745292668,
|
|
"learning_rate": 9.972178779339264e-06,
|
|
"loss": 0.8196,
|
|
"step": 58
|
|
},
|
|
{
|
|
"epoch": 0.06400433927723914,
|
|
"grad_norm": 0.8988095566284031,
|
|
"learning_rate": 9.970294988098452e-06,
|
|
"loss": 0.8144,
|
|
"step": 59
|
|
},
|
|
{
|
|
"epoch": 0.06508915858702286,
|
|
"grad_norm": 0.49627850174956567,
|
|
"learning_rate": 9.968349682209434e-06,
|
|
"loss": 0.8144,
|
|
"step": 60
|
|
},
|
|
{
|
|
"epoch": 0.06508915858702286,
|
|
"eval_loss": 0.8125380277633667,
|
|
"eval_runtime": 109.5146,
|
|
"eval_samples_per_second": 87.057,
|
|
"eval_steps_per_second": 0.685,
|
|
"step": 60
|
|
},
|
|
{
|
|
"epoch": 0.06617397789680657,
|
|
"grad_norm": 0.45744888427391833,
|
|
"learning_rate": 9.966342885748212e-06,
|
|
"loss": 0.8159,
|
|
"step": 61
|
|
},
|
|
{
|
|
"epoch": 0.06725879720659028,
|
|
"grad_norm": 0.7311477252225216,
|
|
"learning_rate": 9.964274623551814e-06,
|
|
"loss": 0.8112,
|
|
"step": 62
|
|
},
|
|
{
|
|
"epoch": 0.068343616516374,
|
|
"grad_norm": 0.7181451822565164,
|
|
"learning_rate": 9.962144921218005e-06,
|
|
"loss": 0.8104,
|
|
"step": 63
|
|
},
|
|
{
|
|
"epoch": 0.06942843582615771,
|
|
"grad_norm": 0.3382680278399959,
|
|
"learning_rate": 9.959953805104953e-06,
|
|
"loss": 0.8075,
|
|
"step": 64
|
|
},
|
|
{
|
|
"epoch": 0.07051325513594142,
|
|
"grad_norm": 0.4704529718605793,
|
|
"learning_rate": 9.957701302330915e-06,
|
|
"loss": 0.812,
|
|
"step": 65
|
|
},
|
|
{
|
|
"epoch": 0.07159807444572514,
|
|
"grad_norm": 0.648293766273129,
|
|
"learning_rate": 9.955387440773902e-06,
|
|
"loss": 0.809,
|
|
"step": 66
|
|
},
|
|
{
|
|
"epoch": 0.07268289375550885,
|
|
"grad_norm": 0.41562985503267486,
|
|
"learning_rate": 9.95301224907132e-06,
|
|
"loss": 0.809,
|
|
"step": 67
|
|
},
|
|
{
|
|
"epoch": 0.07376771306529256,
|
|
"grad_norm": 0.3722271653902924,
|
|
"learning_rate": 9.95057575661963e-06,
|
|
"loss": 0.8011,
|
|
"step": 68
|
|
},
|
|
{
|
|
"epoch": 0.07485253237507627,
|
|
"grad_norm": 0.6079172339989929,
|
|
"learning_rate": 9.948077993573983e-06,
|
|
"loss": 0.8065,
|
|
"step": 69
|
|
},
|
|
{
|
|
"epoch": 0.07593735168485999,
|
|
"grad_norm": 0.5435218074168965,
|
|
"learning_rate": 9.945518990847835e-06,
|
|
"loss": 0.8027,
|
|
"step": 70
|
|
},
|
|
{
|
|
"epoch": 0.07593735168485999,
|
|
"eval_loss": 0.803402841091156,
|
|
"eval_runtime": 108.4779,
|
|
"eval_samples_per_second": 87.889,
|
|
"eval_steps_per_second": 0.691,
|
|
"step": 70
|
|
},
|
|
{
|
|
"epoch": 0.0770221709946437,
|
|
"grad_norm": 0.34820089064729376,
|
|
"learning_rate": 9.942898780112578e-06,
|
|
"loss": 0.8039,
|
|
"step": 71
|
|
},
|
|
{
|
|
"epoch": 0.07810699030442741,
|
|
"grad_norm": 0.44514115903246937,
|
|
"learning_rate": 9.94021739379714e-06,
|
|
"loss": 0.807,
|
|
"step": 72
|
|
},
|
|
{
|
|
"epoch": 0.07919180961421113,
|
|
"grad_norm": 0.5369526957264457,
|
|
"learning_rate": 9.937474865087588e-06,
|
|
"loss": 0.8042,
|
|
"step": 73
|
|
},
|
|
{
|
|
"epoch": 0.08027662892399484,
|
|
"grad_norm": 0.37982616690889814,
|
|
"learning_rate": 9.934671227926714e-06,
|
|
"loss": 0.8015,
|
|
"step": 74
|
|
},
|
|
{
|
|
"epoch": 0.08136144823377855,
|
|
"grad_norm": 0.2774312760519546,
|
|
"learning_rate": 9.931806517013612e-06,
|
|
"loss": 0.7995,
|
|
"step": 75
|
|
},
|
|
{
|
|
"epoch": 0.08244626754356228,
|
|
"grad_norm": 0.4053838968905353,
|
|
"learning_rate": 9.928880767803264e-06,
|
|
"loss": 0.7976,
|
|
"step": 76
|
|
},
|
|
{
|
|
"epoch": 0.083531086853346,
|
|
"grad_norm": 0.37288691711081295,
|
|
"learning_rate": 9.925894016506076e-06,
|
|
"loss": 0.7965,
|
|
"step": 77
|
|
},
|
|
{
|
|
"epoch": 0.08461590616312971,
|
|
"grad_norm": 0.2932226054659,
|
|
"learning_rate": 9.922846300087454e-06,
|
|
"loss": 0.8,
|
|
"step": 78
|
|
},
|
|
{
|
|
"epoch": 0.08570072547291342,
|
|
"grad_norm": 0.3375396039583161,
|
|
"learning_rate": 9.919737656267335e-06,
|
|
"loss": 0.7992,
|
|
"step": 79
|
|
},
|
|
{
|
|
"epoch": 0.08678554478269714,
|
|
"grad_norm": 0.3724197687760549,
|
|
"learning_rate": 9.916568123519713e-06,
|
|
"loss": 0.7954,
|
|
"step": 80
|
|
},
|
|
{
|
|
"epoch": 0.08678554478269714,
|
|
"eval_loss": 0.7967098951339722,
|
|
"eval_runtime": 108.5203,
|
|
"eval_samples_per_second": 87.855,
|
|
"eval_steps_per_second": 0.691,
|
|
"step": 80
|
|
},
|
|
{
|
|
"epoch": 0.08787036409248085,
|
|
"grad_norm": 0.41680195182829916,
|
|
"learning_rate": 9.913337741072183e-06,
|
|
"loss": 0.7949,
|
|
"step": 81
|
|
},
|
|
{
|
|
"epoch": 0.08895518340226456,
|
|
"grad_norm": 0.4680964543106562,
|
|
"learning_rate": 9.910046548905437e-06,
|
|
"loss": 0.796,
|
|
"step": 82
|
|
},
|
|
{
|
|
"epoch": 0.09004000271204828,
|
|
"grad_norm": 0.45610055097374685,
|
|
"learning_rate": 9.906694587752777e-06,
|
|
"loss": 0.7966,
|
|
"step": 83
|
|
},
|
|
{
|
|
"epoch": 0.09112482202183199,
|
|
"grad_norm": 0.31643861291902997,
|
|
"learning_rate": 9.90328189909961e-06,
|
|
"loss": 0.797,
|
|
"step": 84
|
|
},
|
|
{
|
|
"epoch": 0.0922096413316157,
|
|
"grad_norm": 0.26009080853848854,
|
|
"learning_rate": 9.899808525182935e-06,
|
|
"loss": 0.7947,
|
|
"step": 85
|
|
},
|
|
{
|
|
"epoch": 0.09329446064139942,
|
|
"grad_norm": 0.37821058405648994,
|
|
"learning_rate": 9.896274508990818e-06,
|
|
"loss": 0.7957,
|
|
"step": 86
|
|
},
|
|
{
|
|
"epoch": 0.09437927995118313,
|
|
"grad_norm": 0.44173687010200685,
|
|
"learning_rate": 9.892679894261865e-06,
|
|
"loss": 0.795,
|
|
"step": 87
|
|
},
|
|
{
|
|
"epoch": 0.09546409926096684,
|
|
"grad_norm": 0.4715462455517994,
|
|
"learning_rate": 9.889024725484672e-06,
|
|
"loss": 0.7901,
|
|
"step": 88
|
|
},
|
|
{
|
|
"epoch": 0.09654891857075056,
|
|
"grad_norm": 0.4527537880352518,
|
|
"learning_rate": 9.885309047897285e-06,
|
|
"loss": 0.7883,
|
|
"step": 89
|
|
},
|
|
{
|
|
"epoch": 0.09763373788053427,
|
|
"grad_norm": 0.43022666713295693,
|
|
"learning_rate": 9.88153290748663e-06,
|
|
"loss": 0.7895,
|
|
"step": 90
|
|
},
|
|
{
|
|
"epoch": 0.09763373788053427,
|
|
"eval_loss": 0.7914819717407227,
|
|
"eval_runtime": 109.1067,
|
|
"eval_samples_per_second": 87.382,
|
|
"eval_steps_per_second": 0.687,
|
|
"step": 90
|
|
},
|
|
{
|
|
"epoch": 0.09871855719031798,
|
|
"grad_norm": 0.5316622232573952,
|
|
"learning_rate": 9.877696350987954e-06,
|
|
"loss": 0.7918,
|
|
"step": 91
|
|
},
|
|
{
|
|
"epoch": 0.0998033765001017,
|
|
"grad_norm": 0.7272845675318187,
|
|
"learning_rate": 9.873799425884235e-06,
|
|
"loss": 0.7935,
|
|
"step": 92
|
|
},
|
|
{
|
|
"epoch": 0.10088819580988542,
|
|
"grad_norm": 0.9755003013537064,
|
|
"learning_rate": 9.869842180405607e-06,
|
|
"loss": 0.7933,
|
|
"step": 93
|
|
},
|
|
{
|
|
"epoch": 0.10197301511966914,
|
|
"grad_norm": 1.0841458378326259,
|
|
"learning_rate": 9.86582466352875e-06,
|
|
"loss": 0.7916,
|
|
"step": 94
|
|
},
|
|
{
|
|
"epoch": 0.10305783442945285,
|
|
"grad_norm": 0.6302564437210111,
|
|
"learning_rate": 9.861746924976297e-06,
|
|
"loss": 0.791,
|
|
"step": 95
|
|
},
|
|
{
|
|
"epoch": 0.10414265373923656,
|
|
"grad_norm": 0.35908321146074057,
|
|
"learning_rate": 9.857609015216205e-06,
|
|
"loss": 0.7903,
|
|
"step": 96
|
|
},
|
|
{
|
|
"epoch": 0.10522747304902028,
|
|
"grad_norm": 0.6267170848390329,
|
|
"learning_rate": 9.853410985461145e-06,
|
|
"loss": 0.7872,
|
|
"step": 97
|
|
},
|
|
{
|
|
"epoch": 0.10631229235880399,
|
|
"grad_norm": 0.7120302407187888,
|
|
"learning_rate": 9.849152887667855e-06,
|
|
"loss": 0.7891,
|
|
"step": 98
|
|
},
|
|
{
|
|
"epoch": 0.1073971116685877,
|
|
"grad_norm": 0.6858778736543274,
|
|
"learning_rate": 9.844834774536503e-06,
|
|
"loss": 0.7883,
|
|
"step": 99
|
|
},
|
|
{
|
|
"epoch": 0.10848193097837142,
|
|
"grad_norm": 0.4314848431255045,
|
|
"learning_rate": 9.840456699510038e-06,
|
|
"loss": 0.7861,
|
|
"step": 100
|
|
},
|
|
{
|
|
"epoch": 0.10848193097837142,
|
|
"eval_loss": 0.787100076675415,
|
|
"eval_runtime": 109.4114,
|
|
"eval_samples_per_second": 87.139,
|
|
"eval_steps_per_second": 0.685,
|
|
"step": 100
|
|
},
|
|
{
|
|
"epoch": 0.10956675028815513,
|
|
"grad_norm": 0.5194656551626861,
|
|
"learning_rate": 9.836018716773522e-06,
|
|
"loss": 0.7888,
|
|
"step": 101
|
|
},
|
|
{
|
|
"epoch": 0.11065156959793884,
|
|
"grad_norm": 0.742531195114932,
|
|
"learning_rate": 9.831520881253462e-06,
|
|
"loss": 0.7875,
|
|
"step": 102
|
|
},
|
|
{
|
|
"epoch": 0.11173638890772256,
|
|
"grad_norm": 0.4989969979511064,
|
|
"learning_rate": 9.826963248617133e-06,
|
|
"loss": 0.7833,
|
|
"step": 103
|
|
},
|
|
{
|
|
"epoch": 0.11282120821750627,
|
|
"grad_norm": 0.375240047394447,
|
|
"learning_rate": 9.822345875271884e-06,
|
|
"loss": 0.7854,
|
|
"step": 104
|
|
},
|
|
{
|
|
"epoch": 0.11390602752728998,
|
|
"grad_norm": 0.4988303251049989,
|
|
"learning_rate": 9.817668818364441e-06,
|
|
"loss": 0.785,
|
|
"step": 105
|
|
},
|
|
{
|
|
"epoch": 0.1149908468370737,
|
|
"grad_norm": 0.49945188870075524,
|
|
"learning_rate": 9.81293213578021e-06,
|
|
"loss": 0.7875,
|
|
"step": 106
|
|
},
|
|
{
|
|
"epoch": 0.11607566614685741,
|
|
"grad_norm": 0.45320167914195025,
|
|
"learning_rate": 9.808135886142536e-06,
|
|
"loss": 0.7822,
|
|
"step": 107
|
|
},
|
|
{
|
|
"epoch": 0.11716048545664112,
|
|
"grad_norm": 0.3304125239146547,
|
|
"learning_rate": 9.803280128812009e-06,
|
|
"loss": 0.7835,
|
|
"step": 108
|
|
},
|
|
{
|
|
"epoch": 0.11824530476642484,
|
|
"grad_norm": 0.454088367030755,
|
|
"learning_rate": 9.798364923885703e-06,
|
|
"loss": 0.7794,
|
|
"step": 109
|
|
},
|
|
{
|
|
"epoch": 0.11933012407620856,
|
|
"grad_norm": 0.39861269990268244,
|
|
"learning_rate": 9.793390332196448e-06,
|
|
"loss": 0.7823,
|
|
"step": 110
|
|
},
|
|
{
|
|
"epoch": 0.11933012407620856,
|
|
"eval_loss": 0.7828577160835266,
|
|
"eval_runtime": 108.5297,
|
|
"eval_samples_per_second": 87.847,
|
|
"eval_steps_per_second": 0.691,
|
|
"step": 110
|
|
},
|
|
{
|
|
"epoch": 0.12041494338599228,
|
|
"grad_norm": 0.34179794826063836,
|
|
"learning_rate": 9.78835641531207e-06,
|
|
"loss": 0.785,
|
|
"step": 111
|
|
},
|
|
{
|
|
"epoch": 0.12149976269577599,
|
|
"grad_norm": 0.4384145900437222,
|
|
"learning_rate": 9.783263235534632e-06,
|
|
"loss": 0.7827,
|
|
"step": 112
|
|
},
|
|
{
|
|
"epoch": 0.1225845820055597,
|
|
"grad_norm": 0.35050179811766313,
|
|
"learning_rate": 9.778110855899659e-06,
|
|
"loss": 0.7795,
|
|
"step": 113
|
|
},
|
|
{
|
|
"epoch": 0.12366940131534342,
|
|
"grad_norm": 0.37888088278503856,
|
|
"learning_rate": 9.772899340175362e-06,
|
|
"loss": 0.7812,
|
|
"step": 114
|
|
},
|
|
{
|
|
"epoch": 0.12475422062512713,
|
|
"grad_norm": 0.46331100701408845,
|
|
"learning_rate": 9.767628752861848e-06,
|
|
"loss": 0.7795,
|
|
"step": 115
|
|
},
|
|
{
|
|
"epoch": 0.12583903993491083,
|
|
"grad_norm": 0.253077197462135,
|
|
"learning_rate": 9.762299159190322e-06,
|
|
"loss": 0.7794,
|
|
"step": 116
|
|
},
|
|
{
|
|
"epoch": 0.12692385924469454,
|
|
"grad_norm": 0.32841861687248075,
|
|
"learning_rate": 9.756910625122276e-06,
|
|
"loss": 0.7814,
|
|
"step": 117
|
|
},
|
|
{
|
|
"epoch": 0.12800867855447828,
|
|
"grad_norm": 0.3961680026713341,
|
|
"learning_rate": 9.751463217348675e-06,
|
|
"loss": 0.7848,
|
|
"step": 118
|
|
},
|
|
{
|
|
"epoch": 0.129093497864262,
|
|
"grad_norm": 0.361621602491951,
|
|
"learning_rate": 9.745957003289138e-06,
|
|
"loss": 0.7799,
|
|
"step": 119
|
|
},
|
|
{
|
|
"epoch": 0.1301783171740457,
|
|
"grad_norm": 0.45639933190314985,
|
|
"learning_rate": 9.74039205109109e-06,
|
|
"loss": 0.7787,
|
|
"step": 120
|
|
},
|
|
{
|
|
"epoch": 0.1301783171740457,
|
|
"eval_loss": 0.7795460820198059,
|
|
"eval_runtime": 108.5296,
|
|
"eval_samples_per_second": 87.847,
|
|
"eval_steps_per_second": 0.691,
|
|
"step": 120
|
|
},
|
|
{
|
|
"epoch": 0.13126313648382942,
|
|
"grad_norm": 0.4894024734567449,
|
|
"learning_rate": 9.73476842962893e-06,
|
|
"loss": 0.7826,
|
|
"step": 121
|
|
},
|
|
{
|
|
"epoch": 0.13234795579361314,
|
|
"grad_norm": 0.3400805313158744,
|
|
"learning_rate": 9.729086208503174e-06,
|
|
"loss": 0.7774,
|
|
"step": 122
|
|
},
|
|
{
|
|
"epoch": 0.13343277510339685,
|
|
"grad_norm": 0.2807653195556927,
|
|
"learning_rate": 9.723345458039595e-06,
|
|
"loss": 0.777,
|
|
"step": 123
|
|
},
|
|
{
|
|
"epoch": 0.13451759441318056,
|
|
"grad_norm": 0.399127433948202,
|
|
"learning_rate": 9.71754624928835e-06,
|
|
"loss": 0.7791,
|
|
"step": 124
|
|
},
|
|
{
|
|
"epoch": 0.13560241372296428,
|
|
"grad_norm": 0.41268439908392046,
|
|
"learning_rate": 9.711688654023105e-06,
|
|
"loss": 0.7798,
|
|
"step": 125
|
|
},
|
|
{
|
|
"epoch": 0.136687233032748,
|
|
"grad_norm": 0.43996629750095084,
|
|
"learning_rate": 9.705772744740142e-06,
|
|
"loss": 0.7763,
|
|
"step": 126
|
|
},
|
|
{
|
|
"epoch": 0.1377720523425317,
|
|
"grad_norm": 0.4341497876027942,
|
|
"learning_rate": 9.699798594657464e-06,
|
|
"loss": 0.7736,
|
|
"step": 127
|
|
},
|
|
{
|
|
"epoch": 0.13885687165231542,
|
|
"grad_norm": 0.4373015418932958,
|
|
"learning_rate": 9.693766277713893e-06,
|
|
"loss": 0.778,
|
|
"step": 128
|
|
},
|
|
{
|
|
"epoch": 0.13994169096209913,
|
|
"grad_norm": 0.59918719623413,
|
|
"learning_rate": 9.687675868568145e-06,
|
|
"loss": 0.7752,
|
|
"step": 129
|
|
},
|
|
{
|
|
"epoch": 0.14102651027188284,
|
|
"grad_norm": 0.7097994736999107,
|
|
"learning_rate": 9.681527442597916e-06,
|
|
"loss": 0.7769,
|
|
"step": 130
|
|
},
|
|
{
|
|
"epoch": 0.14102651027188284,
|
|
"eval_loss": 0.7773523330688477,
|
|
"eval_runtime": 108.9443,
|
|
"eval_samples_per_second": 87.513,
|
|
"eval_steps_per_second": 0.688,
|
|
"step": 130
|
|
},
|
|
{
|
|
"epoch": 0.14211132958166656,
|
|
"grad_norm": 0.7958519381675093,
|
|
"learning_rate": 9.67532107589894e-06,
|
|
"loss": 0.7789,
|
|
"step": 131
|
|
},
|
|
{
|
|
"epoch": 0.14319614889145027,
|
|
"grad_norm": 0.8068539147321399,
|
|
"learning_rate": 9.66905684528406e-06,
|
|
"loss": 0.7755,
|
|
"step": 132
|
|
},
|
|
{
|
|
"epoch": 0.14428096820123398,
|
|
"grad_norm": 0.5764221421667293,
|
|
"learning_rate": 9.662734828282258e-06,
|
|
"loss": 0.7744,
|
|
"step": 133
|
|
},
|
|
{
|
|
"epoch": 0.1453657875110177,
|
|
"grad_norm": 0.31199910392765157,
|
|
"learning_rate": 9.65635510313772e-06,
|
|
"loss": 0.7768,
|
|
"step": 134
|
|
},
|
|
{
|
|
"epoch": 0.1464506068208014,
|
|
"grad_norm": 0.4041975552185395,
|
|
"learning_rate": 9.649917748808844e-06,
|
|
"loss": 0.7765,
|
|
"step": 135
|
|
},
|
|
{
|
|
"epoch": 0.14753542613058512,
|
|
"grad_norm": 0.6164608497017049,
|
|
"learning_rate": 9.643422844967274e-06,
|
|
"loss": 0.7773,
|
|
"step": 136
|
|
},
|
|
{
|
|
"epoch": 0.14862024544036884,
|
|
"grad_norm": 0.6348166603995771,
|
|
"learning_rate": 9.636870471996923e-06,
|
|
"loss": 0.7706,
|
|
"step": 137
|
|
},
|
|
{
|
|
"epoch": 0.14970506475015255,
|
|
"grad_norm": 0.42070168125876545,
|
|
"learning_rate": 9.630260710992956e-06,
|
|
"loss": 0.774,
|
|
"step": 138
|
|
},
|
|
{
|
|
"epoch": 0.15078988405993626,
|
|
"grad_norm": 0.2647874637161585,
|
|
"learning_rate": 9.623593643760805e-06,
|
|
"loss": 0.7735,
|
|
"step": 139
|
|
},
|
|
{
|
|
"epoch": 0.15187470336971998,
|
|
"grad_norm": 0.45094031514213195,
|
|
"learning_rate": 9.61686935281515e-06,
|
|
"loss": 0.7774,
|
|
"step": 140
|
|
},
|
|
{
|
|
"epoch": 0.15187470336971998,
|
|
"eval_loss": 0.7735901474952698,
|
|
"eval_runtime": 108.4217,
|
|
"eval_samples_per_second": 87.934,
|
|
"eval_steps_per_second": 0.692,
|
|
"step": 140
|
|
},
|
|
{
|
|
"epoch": 0.1529595226795037,
|
|
"grad_norm": 0.5119777928263476,
|
|
"learning_rate": 9.610087921378895e-06,
|
|
"loss": 0.7739,
|
|
"step": 141
|
|
},
|
|
{
|
|
"epoch": 0.1540443419892874,
|
|
"grad_norm": 0.4048675689063343,
|
|
"learning_rate": 9.603249433382145e-06,
|
|
"loss": 0.7779,
|
|
"step": 142
|
|
},
|
|
{
|
|
"epoch": 0.15512916129907112,
|
|
"grad_norm": 0.247541321829258,
|
|
"learning_rate": 9.59635397346116e-06,
|
|
"loss": 0.7724,
|
|
"step": 143
|
|
},
|
|
{
|
|
"epoch": 0.15621398060885483,
|
|
"grad_norm": 0.3537597642765961,
|
|
"learning_rate": 9.589401626957309e-06,
|
|
"loss": 0.7722,
|
|
"step": 144
|
|
},
|
|
{
|
|
"epoch": 0.15729879991863854,
|
|
"grad_norm": 0.4294326059871437,
|
|
"learning_rate": 9.582392479916023e-06,
|
|
"loss": 0.7681,
|
|
"step": 145
|
|
},
|
|
{
|
|
"epoch": 0.15838361922842226,
|
|
"grad_norm": 0.4117348857649006,
|
|
"learning_rate": 9.575326619085713e-06,
|
|
"loss": 0.7758,
|
|
"step": 146
|
|
},
|
|
{
|
|
"epoch": 0.15946843853820597,
|
|
"grad_norm": 0.3422223479430551,
|
|
"learning_rate": 9.568204131916712e-06,
|
|
"loss": 0.7685,
|
|
"step": 147
|
|
},
|
|
{
|
|
"epoch": 0.16055325784798968,
|
|
"grad_norm": 0.2554148075298471,
|
|
"learning_rate": 9.561025106560184e-06,
|
|
"loss": 0.7728,
|
|
"step": 148
|
|
},
|
|
{
|
|
"epoch": 0.1616380771577734,
|
|
"grad_norm": 0.29437098480907653,
|
|
"learning_rate": 9.553789631867039e-06,
|
|
"loss": 0.77,
|
|
"step": 149
|
|
},
|
|
{
|
|
"epoch": 0.1627228964675571,
|
|
"grad_norm": 0.34279254034717055,
|
|
"learning_rate": 9.546497797386824e-06,
|
|
"loss": 0.7701,
|
|
"step": 150
|
|
},
|
|
{
|
|
"epoch": 0.1627228964675571,
|
|
"eval_loss": 0.7706267237663269,
|
|
"eval_runtime": 108.6103,
|
|
"eval_samples_per_second": 87.782,
|
|
"eval_steps_per_second": 0.691,
|
|
"step": 150
|
|
},
|
|
{
|
|
"epoch": 0.16380771577734082,
|
|
"grad_norm": 0.3768245607446824,
|
|
"learning_rate": 9.539149693366628e-06,
|
|
"loss": 0.7699,
|
|
"step": 151
|
|
},
|
|
{
|
|
"epoch": 0.16489253508712456,
|
|
"grad_norm": 0.2795450346743165,
|
|
"learning_rate": 9.531745410749946e-06,
|
|
"loss": 0.7719,
|
|
"step": 152
|
|
},
|
|
{
|
|
"epoch": 0.16597735439690828,
|
|
"grad_norm": 0.21565184349732974,
|
|
"learning_rate": 9.524285041175578e-06,
|
|
"loss": 0.7709,
|
|
"step": 153
|
|
},
|
|
{
|
|
"epoch": 0.167062173706692,
|
|
"grad_norm": 0.27917354068385714,
|
|
"learning_rate": 9.516768676976476e-06,
|
|
"loss": 0.7655,
|
|
"step": 154
|
|
},
|
|
{
|
|
"epoch": 0.1681469930164757,
|
|
"grad_norm": 0.36818858829997037,
|
|
"learning_rate": 9.509196411178605e-06,
|
|
"loss": 0.7702,
|
|
"step": 155
|
|
},
|
|
{
|
|
"epoch": 0.16923181232625942,
|
|
"grad_norm": 0.3931634948674058,
|
|
"learning_rate": 9.501568337499798e-06,
|
|
"loss": 0.7698,
|
|
"step": 156
|
|
},
|
|
{
|
|
"epoch": 0.17031663163604313,
|
|
"grad_norm": 0.3711182444316531,
|
|
"learning_rate": 9.493884550348589e-06,
|
|
"loss": 0.7689,
|
|
"step": 157
|
|
},
|
|
{
|
|
"epoch": 0.17140145094582684,
|
|
"grad_norm": 0.34420286948578516,
|
|
"learning_rate": 9.48614514482305e-06,
|
|
"loss": 0.7678,
|
|
"step": 158
|
|
},
|
|
{
|
|
"epoch": 0.17248627025561056,
|
|
"grad_norm": 0.36595434050790837,
|
|
"learning_rate": 9.478350216709609e-06,
|
|
"loss": 0.7692,
|
|
"step": 159
|
|
},
|
|
{
|
|
"epoch": 0.17357108956539427,
|
|
"grad_norm": 0.46016642623218235,
|
|
"learning_rate": 9.470499862481867e-06,
|
|
"loss": 0.7691,
|
|
"step": 160
|
|
},
|
|
{
|
|
"epoch": 0.17357108956539427,
|
|
"eval_loss": 0.7686050534248352,
|
|
"eval_runtime": 108.3988,
|
|
"eval_samples_per_second": 87.953,
|
|
"eval_steps_per_second": 0.692,
|
|
"step": 160
|
|
},
|
|
{
|
|
"epoch": 0.17465590887517798,
|
|
"grad_norm": 0.5189068589203444,
|
|
"learning_rate": 9.462594179299408e-06,
|
|
"loss": 0.7691,
|
|
"step": 161
|
|
},
|
|
{
|
|
"epoch": 0.1757407281849617,
|
|
"grad_norm": 0.5037770127077009,
|
|
"learning_rate": 9.454633265006585e-06,
|
|
"loss": 0.7718,
|
|
"step": 162
|
|
},
|
|
{
|
|
"epoch": 0.1768255474947454,
|
|
"grad_norm": 0.4262772872702808,
|
|
"learning_rate": 9.446617218131326e-06,
|
|
"loss": 0.7688,
|
|
"step": 163
|
|
},
|
|
{
|
|
"epoch": 0.17791036680452912,
|
|
"grad_norm": 0.3030752504131984,
|
|
"learning_rate": 9.438546137883898e-06,
|
|
"loss": 0.7642,
|
|
"step": 164
|
|
},
|
|
{
|
|
"epoch": 0.17899518611431284,
|
|
"grad_norm": 0.28877592332321944,
|
|
"learning_rate": 9.430420124155687e-06,
|
|
"loss": 0.7672,
|
|
"step": 165
|
|
},
|
|
{
|
|
"epoch": 0.18008000542409655,
|
|
"grad_norm": 0.3135773421625362,
|
|
"learning_rate": 9.422239277517964e-06,
|
|
"loss": 0.7685,
|
|
"step": 166
|
|
},
|
|
{
|
|
"epoch": 0.18116482473388026,
|
|
"grad_norm": 0.3681496362575959,
|
|
"learning_rate": 9.414003699220636e-06,
|
|
"loss": 0.7666,
|
|
"step": 167
|
|
},
|
|
{
|
|
"epoch": 0.18224964404366398,
|
|
"grad_norm": 0.33019072670103566,
|
|
"learning_rate": 9.405713491190992e-06,
|
|
"loss": 0.7689,
|
|
"step": 168
|
|
},
|
|
{
|
|
"epoch": 0.1833344633534477,
|
|
"grad_norm": 0.24672102847526706,
|
|
"learning_rate": 9.397368756032445e-06,
|
|
"loss": 0.7671,
|
|
"step": 169
|
|
},
|
|
{
|
|
"epoch": 0.1844192826632314,
|
|
"grad_norm": 0.3061688738027205,
|
|
"learning_rate": 9.388969597023265e-06,
|
|
"loss": 0.767,
|
|
"step": 170
|
|
},
|
|
{
|
|
"epoch": 0.1844192826632314,
|
|
"eval_loss": 0.7660329341888428,
|
|
"eval_runtime": 108.314,
|
|
"eval_samples_per_second": 88.022,
|
|
"eval_steps_per_second": 0.692,
|
|
"step": 170
|
|
},
|
|
{
|
|
"epoch": 0.18550410197301512,
|
|
"grad_norm": 0.41309286266345396,
|
|
"learning_rate": 9.380516118115287e-06,
|
|
"loss": 0.7644,
|
|
"step": 171
|
|
},
|
|
{
|
|
"epoch": 0.18658892128279883,
|
|
"grad_norm": 0.43250642443620424,
|
|
"learning_rate": 9.37200842393264e-06,
|
|
"loss": 0.7671,
|
|
"step": 172
|
|
},
|
|
{
|
|
"epoch": 0.18767374059258254,
|
|
"grad_norm": 0.36310207799309735,
|
|
"learning_rate": 9.363446619770452e-06,
|
|
"loss": 0.767,
|
|
"step": 173
|
|
},
|
|
{
|
|
"epoch": 0.18875855990236626,
|
|
"grad_norm": 0.29200489152969605,
|
|
"learning_rate": 9.354830811593527e-06,
|
|
"loss": 0.7683,
|
|
"step": 174
|
|
},
|
|
{
|
|
"epoch": 0.18984337921214997,
|
|
"grad_norm": 0.31700471252436824,
|
|
"learning_rate": 9.346161106035056e-06,
|
|
"loss": 0.7662,
|
|
"step": 175
|
|
},
|
|
{
|
|
"epoch": 0.19092819852193368,
|
|
"grad_norm": 0.4123517398509975,
|
|
"learning_rate": 9.337437610395292e-06,
|
|
"loss": 0.7682,
|
|
"step": 176
|
|
},
|
|
{
|
|
"epoch": 0.1920130178317174,
|
|
"grad_norm": 0.4461455348284238,
|
|
"learning_rate": 9.328660432640211e-06,
|
|
"loss": 0.7682,
|
|
"step": 177
|
|
},
|
|
{
|
|
"epoch": 0.1930978371415011,
|
|
"grad_norm": 0.41857803905013113,
|
|
"learning_rate": 9.319829681400185e-06,
|
|
"loss": 0.7672,
|
|
"step": 178
|
|
},
|
|
{
|
|
"epoch": 0.19418265645128482,
|
|
"grad_norm": 0.3544867837407869,
|
|
"learning_rate": 9.31094546596864e-06,
|
|
"loss": 0.7624,
|
|
"step": 179
|
|
},
|
|
{
|
|
"epoch": 0.19526747576106854,
|
|
"grad_norm": 0.26138631108200805,
|
|
"learning_rate": 9.302007896300697e-06,
|
|
"loss": 0.7617,
|
|
"step": 180
|
|
},
|
|
{
|
|
"epoch": 0.19526747576106854,
|
|
"eval_loss": 0.7637682557106018,
|
|
"eval_runtime": 108.4177,
|
|
"eval_samples_per_second": 87.938,
|
|
"eval_steps_per_second": 0.692,
|
|
"step": 180
|
|
},
|
|
{
|
|
"epoch": 0.19635229507085225,
|
|
"grad_norm": 0.3688297227057638,
|
|
"learning_rate": 9.293017083011814e-06,
|
|
"loss": 0.7613,
|
|
"step": 181
|
|
},
|
|
{
|
|
"epoch": 0.19743711438063596,
|
|
"grad_norm": 0.5028925982918173,
|
|
"learning_rate": 9.283973137376414e-06,
|
|
"loss": 0.7661,
|
|
"step": 182
|
|
},
|
|
{
|
|
"epoch": 0.19852193369041968,
|
|
"grad_norm": 0.5024698547461525,
|
|
"learning_rate": 9.274876171326514e-06,
|
|
"loss": 0.7638,
|
|
"step": 183
|
|
},
|
|
{
|
|
"epoch": 0.1996067530002034,
|
|
"grad_norm": 0.4288303633868785,
|
|
"learning_rate": 9.265726297450332e-06,
|
|
"loss": 0.7622,
|
|
"step": 184
|
|
},
|
|
{
|
|
"epoch": 0.2006915723099871,
|
|
"grad_norm": 0.3731677319203024,
|
|
"learning_rate": 9.256523628990903e-06,
|
|
"loss": 0.7632,
|
|
"step": 185
|
|
},
|
|
{
|
|
"epoch": 0.20177639161977085,
|
|
"grad_norm": 0.4824027276068218,
|
|
"learning_rate": 9.247268279844666e-06,
|
|
"loss": 0.7609,
|
|
"step": 186
|
|
},
|
|
{
|
|
"epoch": 0.20286121092955456,
|
|
"grad_norm": 0.6175104754218527,
|
|
"learning_rate": 9.237960364560063e-06,
|
|
"loss": 0.7649,
|
|
"step": 187
|
|
},
|
|
{
|
|
"epoch": 0.20394603023933827,
|
|
"grad_norm": 0.6604414879657724,
|
|
"learning_rate": 9.228599998336119e-06,
|
|
"loss": 0.762,
|
|
"step": 188
|
|
},
|
|
{
|
|
"epoch": 0.20503084954912199,
|
|
"grad_norm": 0.5823286121682828,
|
|
"learning_rate": 9.219187297021015e-06,
|
|
"loss": 0.7632,
|
|
"step": 189
|
|
},
|
|
{
|
|
"epoch": 0.2061156688589057,
|
|
"grad_norm": 0.48249169498286976,
|
|
"learning_rate": 9.209722377110657e-06,
|
|
"loss": 0.7593,
|
|
"step": 190
|
|
},
|
|
{
|
|
"epoch": 0.2061156688589057,
|
|
"eval_loss": 0.761928915977478,
|
|
"eval_runtime": 109.1338,
|
|
"eval_samples_per_second": 87.361,
|
|
"eval_steps_per_second": 0.687,
|
|
"step": 190
|
|
},
|
|
{
|
|
"epoch": 0.2072004881686894,
|
|
"grad_norm": 0.47471914524239467,
|
|
"learning_rate": 9.200205355747228e-06,
|
|
"loss": 0.7639,
|
|
"step": 191
|
|
},
|
|
{
|
|
"epoch": 0.20828530747847313,
|
|
"grad_norm": 0.4955570563237179,
|
|
"learning_rate": 9.190636350717747e-06,
|
|
"loss": 0.7576,
|
|
"step": 192
|
|
},
|
|
{
|
|
"epoch": 0.20937012678825684,
|
|
"grad_norm": 0.42730203058077343,
|
|
"learning_rate": 9.181015480452607e-06,
|
|
"loss": 0.7598,
|
|
"step": 193
|
|
},
|
|
{
|
|
"epoch": 0.21045494609804055,
|
|
"grad_norm": 0.3677256683576732,
|
|
"learning_rate": 9.171342864024103e-06,
|
|
"loss": 0.7609,
|
|
"step": 194
|
|
},
|
|
{
|
|
"epoch": 0.21153976540782427,
|
|
"grad_norm": 0.48632780295999734,
|
|
"learning_rate": 9.161618621144967e-06,
|
|
"loss": 0.759,
|
|
"step": 195
|
|
},
|
|
{
|
|
"epoch": 0.21262458471760798,
|
|
"grad_norm": 0.6412685524464289,
|
|
"learning_rate": 9.15184287216689e-06,
|
|
"loss": 0.7615,
|
|
"step": 196
|
|
},
|
|
{
|
|
"epoch": 0.2137094040273917,
|
|
"grad_norm": 0.6815542331152938,
|
|
"learning_rate": 9.142015738079017e-06,
|
|
"loss": 0.7614,
|
|
"step": 197
|
|
},
|
|
{
|
|
"epoch": 0.2147942233371754,
|
|
"grad_norm": 0.6199432155090734,
|
|
"learning_rate": 9.132137340506464e-06,
|
|
"loss": 0.7605,
|
|
"step": 198
|
|
},
|
|
{
|
|
"epoch": 0.21587904264695912,
|
|
"grad_norm": 0.5079916745649751,
|
|
"learning_rate": 9.122207801708802e-06,
|
|
"loss": 0.7604,
|
|
"step": 199
|
|
},
|
|
{
|
|
"epoch": 0.21696386195674283,
|
|
"grad_norm": 0.4210240875496384,
|
|
"learning_rate": 9.112227244578557e-06,
|
|
"loss": 0.7601,
|
|
"step": 200
|
|
},
|
|
{
|
|
"epoch": 0.21696386195674283,
|
|
"eval_loss": 0.759892463684082,
|
|
"eval_runtime": 108.6381,
|
|
"eval_samples_per_second": 87.759,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 200
|
|
},
|
|
{
|
|
"epoch": 0.21804868126652655,
|
|
"grad_norm": 0.3367318143142601,
|
|
"learning_rate": 9.102195792639677e-06,
|
|
"loss": 0.7602,
|
|
"step": 201
|
|
},
|
|
{
|
|
"epoch": 0.21913350057631026,
|
|
"grad_norm": 0.3206946440980973,
|
|
"learning_rate": 9.092113570046005e-06,
|
|
"loss": 0.7619,
|
|
"step": 202
|
|
},
|
|
{
|
|
"epoch": 0.22021831988609397,
|
|
"grad_norm": 0.4236827917274014,
|
|
"learning_rate": 9.08198070157975e-06,
|
|
"loss": 0.7623,
|
|
"step": 203
|
|
},
|
|
{
|
|
"epoch": 0.22130313919587769,
|
|
"grad_norm": 0.4744063310413663,
|
|
"learning_rate": 9.071797312649934e-06,
|
|
"loss": 0.7639,
|
|
"step": 204
|
|
},
|
|
{
|
|
"epoch": 0.2223879585056614,
|
|
"grad_norm": 0.39602356930919735,
|
|
"learning_rate": 9.061563529290845e-06,
|
|
"loss": 0.7629,
|
|
"step": 205
|
|
},
|
|
{
|
|
"epoch": 0.2234727778154451,
|
|
"grad_norm": 0.3911558770249337,
|
|
"learning_rate": 9.051279478160475e-06,
|
|
"loss": 0.7609,
|
|
"step": 206
|
|
},
|
|
{
|
|
"epoch": 0.22455759712522882,
|
|
"grad_norm": 0.4187149135112974,
|
|
"learning_rate": 9.040945286538954e-06,
|
|
"loss": 0.7558,
|
|
"step": 207
|
|
},
|
|
{
|
|
"epoch": 0.22564241643501254,
|
|
"grad_norm": 0.3580839570301583,
|
|
"learning_rate": 9.03056108232697e-06,
|
|
"loss": 0.757,
|
|
"step": 208
|
|
},
|
|
{
|
|
"epoch": 0.22672723574479625,
|
|
"grad_norm": 0.3169587609709413,
|
|
"learning_rate": 9.020126994044194e-06,
|
|
"loss": 0.7595,
|
|
"step": 209
|
|
},
|
|
{
|
|
"epoch": 0.22781205505457996,
|
|
"grad_norm": 0.3723856557080199,
|
|
"learning_rate": 9.009643150827683e-06,
|
|
"loss": 0.7571,
|
|
"step": 210
|
|
},
|
|
{
|
|
"epoch": 0.22781205505457996,
|
|
"eval_loss": 0.7582467794418335,
|
|
"eval_runtime": 108.9754,
|
|
"eval_samples_per_second": 87.488,
|
|
"eval_steps_per_second": 0.688,
|
|
"step": 210
|
|
},
|
|
{
|
|
"epoch": 0.22889687436436368,
|
|
"grad_norm": 0.41041293164648146,
|
|
"learning_rate": 8.999109682430288e-06,
|
|
"loss": 0.7587,
|
|
"step": 211
|
|
},
|
|
{
|
|
"epoch": 0.2299816936741474,
|
|
"grad_norm": 0.36496514762357657,
|
|
"learning_rate": 8.988526719219035e-06,
|
|
"loss": 0.7553,
|
|
"step": 212
|
|
},
|
|
{
|
|
"epoch": 0.2310665129839311,
|
|
"grad_norm": 0.2752348504987265,
|
|
"learning_rate": 8.977894392173527e-06,
|
|
"loss": 0.7536,
|
|
"step": 213
|
|
},
|
|
{
|
|
"epoch": 0.23215133229371482,
|
|
"grad_norm": 0.28021751119324473,
|
|
"learning_rate": 8.96721283288432e-06,
|
|
"loss": 0.7593,
|
|
"step": 214
|
|
},
|
|
{
|
|
"epoch": 0.23323615160349853,
|
|
"grad_norm": 0.36712529096201973,
|
|
"learning_rate": 8.956482173551281e-06,
|
|
"loss": 0.7576,
|
|
"step": 215
|
|
},
|
|
{
|
|
"epoch": 0.23432097091328224,
|
|
"grad_norm": 0.4055048594091963,
|
|
"learning_rate": 8.94570254698197e-06,
|
|
"loss": 0.7572,
|
|
"step": 216
|
|
},
|
|
{
|
|
"epoch": 0.23540579022306596,
|
|
"grad_norm": 0.39162153231191227,
|
|
"learning_rate": 8.934874086589981e-06,
|
|
"loss": 0.7584,
|
|
"step": 217
|
|
},
|
|
{
|
|
"epoch": 0.23649060953284967,
|
|
"grad_norm": 0.36419982640253645,
|
|
"learning_rate": 8.923996926393306e-06,
|
|
"loss": 0.7597,
|
|
"step": 218
|
|
},
|
|
{
|
|
"epoch": 0.2375754288426334,
|
|
"grad_norm": 0.39668898762559396,
|
|
"learning_rate": 8.91307120101266e-06,
|
|
"loss": 0.7565,
|
|
"step": 219
|
|
},
|
|
{
|
|
"epoch": 0.23866024815241713,
|
|
"grad_norm": 0.3734661079749432,
|
|
"learning_rate": 8.90209704566983e-06,
|
|
"loss": 0.7575,
|
|
"step": 220
|
|
},
|
|
{
|
|
"epoch": 0.23866024815241713,
|
|
"eval_loss": 0.7563009262084961,
|
|
"eval_runtime": 108.6253,
|
|
"eval_samples_per_second": 87.77,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 220
|
|
},
|
|
{
|
|
"epoch": 0.23974506746220084,
|
|
"grad_norm": 0.3057044185692497,
|
|
"learning_rate": 8.891074596185987e-06,
|
|
"loss": 0.7527,
|
|
"step": 221
|
|
},
|
|
{
|
|
"epoch": 0.24082988677198455,
|
|
"grad_norm": 0.309315871824788,
|
|
"learning_rate": 8.880003988980019e-06,
|
|
"loss": 0.7531,
|
|
"step": 222
|
|
},
|
|
{
|
|
"epoch": 0.24191470608176827,
|
|
"grad_norm": 0.2937097460831741,
|
|
"learning_rate": 8.868885361066835e-06,
|
|
"loss": 0.7593,
|
|
"step": 223
|
|
},
|
|
{
|
|
"epoch": 0.24299952539155198,
|
|
"grad_norm": 0.27254631539828456,
|
|
"learning_rate": 8.857718850055663e-06,
|
|
"loss": 0.7536,
|
|
"step": 224
|
|
},
|
|
{
|
|
"epoch": 0.2440843447013357,
|
|
"grad_norm": 0.24538335255869048,
|
|
"learning_rate": 8.846504594148366e-06,
|
|
"loss": 0.755,
|
|
"step": 225
|
|
},
|
|
{
|
|
"epoch": 0.2451691640111194,
|
|
"grad_norm": 0.2910298293243265,
|
|
"learning_rate": 8.83524273213771e-06,
|
|
"loss": 0.7534,
|
|
"step": 226
|
|
},
|
|
{
|
|
"epoch": 0.24625398332090312,
|
|
"grad_norm": 0.395236316918257,
|
|
"learning_rate": 8.82393340340566e-06,
|
|
"loss": 0.7534,
|
|
"step": 227
|
|
},
|
|
{
|
|
"epoch": 0.24733880263068683,
|
|
"grad_norm": 0.42876366746833755,
|
|
"learning_rate": 8.812576747921653e-06,
|
|
"loss": 0.7576,
|
|
"step": 228
|
|
},
|
|
{
|
|
"epoch": 0.24842362194047055,
|
|
"grad_norm": 0.3816843148969155,
|
|
"learning_rate": 8.801172906240857e-06,
|
|
"loss": 0.7493,
|
|
"step": 229
|
|
},
|
|
{
|
|
"epoch": 0.24950844125025426,
|
|
"grad_norm": 0.3453411422665227,
|
|
"learning_rate": 8.789722019502444e-06,
|
|
"loss": 0.7522,
|
|
"step": 230
|
|
},
|
|
{
|
|
"epoch": 0.24950844125025426,
|
|
"eval_loss": 0.7549680471420288,
|
|
"eval_runtime": 108.6791,
|
|
"eval_samples_per_second": 87.726,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 230
|
|
},
|
|
{
|
|
"epoch": 0.25059326056003794,
|
|
"grad_norm": 0.33435596788854244,
|
|
"learning_rate": 8.778224229427836e-06,
|
|
"loss": 0.7541,
|
|
"step": 231
|
|
},
|
|
{
|
|
"epoch": 0.25167807986982166,
|
|
"grad_norm": 0.27760764351944195,
|
|
"learning_rate": 8.76667967831895e-06,
|
|
"loss": 0.7537,
|
|
"step": 232
|
|
},
|
|
{
|
|
"epoch": 0.25276289917960537,
|
|
"grad_norm": 0.24615817421935557,
|
|
"learning_rate": 8.75508850905644e-06,
|
|
"loss": 0.7516,
|
|
"step": 233
|
|
},
|
|
{
|
|
"epoch": 0.2538477184893891,
|
|
"grad_norm": 0.21303754681718962,
|
|
"learning_rate": 8.743450865097929e-06,
|
|
"loss": 0.7552,
|
|
"step": 234
|
|
},
|
|
{
|
|
"epoch": 0.2549325377991728,
|
|
"grad_norm": 0.2769628927578121,
|
|
"learning_rate": 8.731766890476232e-06,
|
|
"loss": 0.7551,
|
|
"step": 235
|
|
},
|
|
{
|
|
"epoch": 0.25601735710895657,
|
|
"grad_norm": 0.2976968148335059,
|
|
"learning_rate": 8.72003672979757e-06,
|
|
"loss": 0.7535,
|
|
"step": 236
|
|
},
|
|
{
|
|
"epoch": 0.2571021764187403,
|
|
"grad_norm": 0.3214565474804767,
|
|
"learning_rate": 8.708260528239788e-06,
|
|
"loss": 0.7504,
|
|
"step": 237
|
|
},
|
|
{
|
|
"epoch": 0.258186995728524,
|
|
"grad_norm": 0.29994958657525217,
|
|
"learning_rate": 8.696438431550553e-06,
|
|
"loss": 0.7552,
|
|
"step": 238
|
|
},
|
|
{
|
|
"epoch": 0.2592718150383077,
|
|
"grad_norm": 0.2535746893639996,
|
|
"learning_rate": 8.68457058604555e-06,
|
|
"loss": 0.7517,
|
|
"step": 239
|
|
},
|
|
{
|
|
"epoch": 0.2603566343480914,
|
|
"grad_norm": 0.2414834021249483,
|
|
"learning_rate": 8.672657138606672e-06,
|
|
"loss": 0.7526,
|
|
"step": 240
|
|
},
|
|
{
|
|
"epoch": 0.2603566343480914,
|
|
"eval_loss": 0.7532132863998413,
|
|
"eval_runtime": 108.3941,
|
|
"eval_samples_per_second": 87.957,
|
|
"eval_steps_per_second": 0.692,
|
|
"step": 240
|
|
},
|
|
{
|
|
"epoch": 0.26144145365787513,
|
|
"grad_norm": 0.2593021989688809,
|
|
"learning_rate": 8.660698236680205e-06,
|
|
"loss": 0.7503,
|
|
"step": 241
|
|
},
|
|
{
|
|
"epoch": 0.26252627296765885,
|
|
"grad_norm": 0.2869190384299556,
|
|
"learning_rate": 8.648694028274998e-06,
|
|
"loss": 0.7548,
|
|
"step": 242
|
|
},
|
|
{
|
|
"epoch": 0.26361109227744256,
|
|
"grad_norm": 0.28923973171158157,
|
|
"learning_rate": 8.636644661960634e-06,
|
|
"loss": 0.7552,
|
|
"step": 243
|
|
},
|
|
{
|
|
"epoch": 0.2646959115872263,
|
|
"grad_norm": 0.29578832607382094,
|
|
"learning_rate": 8.624550286865592e-06,
|
|
"loss": 0.7507,
|
|
"step": 244
|
|
},
|
|
{
|
|
"epoch": 0.26578073089701,
|
|
"grad_norm": 0.2647016805158304,
|
|
"learning_rate": 8.6124110526754e-06,
|
|
"loss": 0.7511,
|
|
"step": 245
|
|
},
|
|
{
|
|
"epoch": 0.2668655502067937,
|
|
"grad_norm": 0.3398998370526537,
|
|
"learning_rate": 8.600227109630785e-06,
|
|
"loss": 0.7546,
|
|
"step": 246
|
|
},
|
|
{
|
|
"epoch": 0.2679503695165774,
|
|
"grad_norm": 0.37152897309018407,
|
|
"learning_rate": 8.587998608525806e-06,
|
|
"loss": 0.7516,
|
|
"step": 247
|
|
},
|
|
{
|
|
"epoch": 0.2690351888263611,
|
|
"grad_norm": 0.37464147111710416,
|
|
"learning_rate": 8.575725700705995e-06,
|
|
"loss": 0.7526,
|
|
"step": 248
|
|
},
|
|
{
|
|
"epoch": 0.27012000813614484,
|
|
"grad_norm": 0.27184480840669567,
|
|
"learning_rate": 8.563408538066486e-06,
|
|
"loss": 0.7522,
|
|
"step": 249
|
|
},
|
|
{
|
|
"epoch": 0.27120482744592855,
|
|
"grad_norm": 0.24275358480991396,
|
|
"learning_rate": 8.551047273050126e-06,
|
|
"loss": 0.7543,
|
|
"step": 250
|
|
},
|
|
{
|
|
"epoch": 0.27120482744592855,
|
|
"eval_loss": 0.7519296407699585,
|
|
"eval_runtime": 108.5466,
|
|
"eval_samples_per_second": 87.833,
|
|
"eval_steps_per_second": 0.691,
|
|
"step": 250
|
|
},
|
|
{
|
|
"epoch": 0.27228964675571227,
|
|
"grad_norm": 0.27384476452821827,
|
|
"learning_rate": 8.538642058645595e-06,
|
|
"loss": 0.7522,
|
|
"step": 251
|
|
},
|
|
{
|
|
"epoch": 0.273374466065496,
|
|
"grad_norm": 0.32879473710949936,
|
|
"learning_rate": 8.52619304838551e-06,
|
|
"loss": 0.7509,
|
|
"step": 252
|
|
},
|
|
{
|
|
"epoch": 0.2744592853752797,
|
|
"grad_norm": 0.31324825847904214,
|
|
"learning_rate": 8.513700396344527e-06,
|
|
"loss": 0.752,
|
|
"step": 253
|
|
},
|
|
{
|
|
"epoch": 0.2755441046850634,
|
|
"grad_norm": 0.2735265150562634,
|
|
"learning_rate": 8.501164257137431e-06,
|
|
"loss": 0.7505,
|
|
"step": 254
|
|
},
|
|
{
|
|
"epoch": 0.2766289239948471,
|
|
"grad_norm": 0.218751699494315,
|
|
"learning_rate": 8.488584785917226e-06,
|
|
"loss": 0.7512,
|
|
"step": 255
|
|
},
|
|
{
|
|
"epoch": 0.27771374330463083,
|
|
"grad_norm": 0.2510317385657662,
|
|
"learning_rate": 8.475962138373212e-06,
|
|
"loss": 0.7519,
|
|
"step": 256
|
|
},
|
|
{
|
|
"epoch": 0.27879856261441455,
|
|
"grad_norm": 0.2767227636063936,
|
|
"learning_rate": 8.463296470729058e-06,
|
|
"loss": 0.751,
|
|
"step": 257
|
|
},
|
|
{
|
|
"epoch": 0.27988338192419826,
|
|
"grad_norm": 0.26977318630564684,
|
|
"learning_rate": 8.45058793974087e-06,
|
|
"loss": 0.7488,
|
|
"step": 258
|
|
},
|
|
{
|
|
"epoch": 0.280968201233982,
|
|
"grad_norm": 0.26313218737676025,
|
|
"learning_rate": 8.437836702695253e-06,
|
|
"loss": 0.7544,
|
|
"step": 259
|
|
},
|
|
{
|
|
"epoch": 0.2820530205437657,
|
|
"grad_norm": 0.24587935883855758,
|
|
"learning_rate": 8.425042917407358e-06,
|
|
"loss": 0.7493,
|
|
"step": 260
|
|
},
|
|
{
|
|
"epoch": 0.2820530205437657,
|
|
"eval_loss": 0.7504924535751343,
|
|
"eval_runtime": 108.6591,
|
|
"eval_samples_per_second": 87.742,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 260
|
|
},
|
|
{
|
|
"epoch": 0.2831378398535494,
|
|
"grad_norm": 0.24683743750116008,
|
|
"learning_rate": 8.41220674221893e-06,
|
|
"loss": 0.7533,
|
|
"step": 261
|
|
},
|
|
{
|
|
"epoch": 0.2842226591633331,
|
|
"grad_norm": 0.2871619714152881,
|
|
"learning_rate": 8.399328335996362e-06,
|
|
"loss": 0.7512,
|
|
"step": 262
|
|
},
|
|
{
|
|
"epoch": 0.2853074784731168,
|
|
"grad_norm": 0.2849411778950338,
|
|
"learning_rate": 8.386407858128707e-06,
|
|
"loss": 0.7511,
|
|
"step": 263
|
|
},
|
|
{
|
|
"epoch": 0.28639229778290054,
|
|
"grad_norm": 0.25799250994684664,
|
|
"learning_rate": 8.373445468525719e-06,
|
|
"loss": 0.7496,
|
|
"step": 264
|
|
},
|
|
{
|
|
"epoch": 0.28747711709268425,
|
|
"grad_norm": 0.29005054046758605,
|
|
"learning_rate": 8.360441327615868e-06,
|
|
"loss": 0.7502,
|
|
"step": 265
|
|
},
|
|
{
|
|
"epoch": 0.28856193640246797,
|
|
"grad_norm": 0.23828656171621873,
|
|
"learning_rate": 8.347395596344365e-06,
|
|
"loss": 0.7473,
|
|
"step": 266
|
|
},
|
|
{
|
|
"epoch": 0.2896467557122517,
|
|
"grad_norm": 0.2733474835529245,
|
|
"learning_rate": 8.334308436171159e-06,
|
|
"loss": 0.7504,
|
|
"step": 267
|
|
},
|
|
{
|
|
"epoch": 0.2907315750220354,
|
|
"grad_norm": 0.3143040804960139,
|
|
"learning_rate": 8.321180009068937e-06,
|
|
"loss": 0.7517,
|
|
"step": 268
|
|
},
|
|
{
|
|
"epoch": 0.2918163943318191,
|
|
"grad_norm": 0.3803037623534588,
|
|
"learning_rate": 8.30801047752113e-06,
|
|
"loss": 0.7488,
|
|
"step": 269
|
|
},
|
|
{
|
|
"epoch": 0.2929012136416028,
|
|
"grad_norm": 0.48167755065397216,
|
|
"learning_rate": 8.294800004519895e-06,
|
|
"loss": 0.7493,
|
|
"step": 270
|
|
},
|
|
{
|
|
"epoch": 0.2929012136416028,
|
|
"eval_loss": 0.7500367760658264,
|
|
"eval_runtime": 108.548,
|
|
"eval_samples_per_second": 87.832,
|
|
"eval_steps_per_second": 0.691,
|
|
"step": 270
|
|
},
|
|
{
|
|
"epoch": 0.29398603295138653,
|
|
"grad_norm": 0.5881094712735191,
|
|
"learning_rate": 8.281548753564101e-06,
|
|
"loss": 0.7497,
|
|
"step": 271
|
|
},
|
|
{
|
|
"epoch": 0.29507085226117025,
|
|
"grad_norm": 0.7133828038348895,
|
|
"learning_rate": 8.2682568886573e-06,
|
|
"loss": 0.7482,
|
|
"step": 272
|
|
},
|
|
{
|
|
"epoch": 0.29615567157095396,
|
|
"grad_norm": 0.7880504107407628,
|
|
"learning_rate": 8.254924574305698e-06,
|
|
"loss": 0.7509,
|
|
"step": 273
|
|
},
|
|
{
|
|
"epoch": 0.2972404908807377,
|
|
"grad_norm": 0.6497699070105365,
|
|
"learning_rate": 8.241551975516133e-06,
|
|
"loss": 0.7512,
|
|
"step": 274
|
|
},
|
|
{
|
|
"epoch": 0.2983253101905214,
|
|
"grad_norm": 0.3527254687665259,
|
|
"learning_rate": 8.228139257794012e-06,
|
|
"loss": 0.75,
|
|
"step": 275
|
|
},
|
|
{
|
|
"epoch": 0.2994101295003051,
|
|
"grad_norm": 0.3146168115606964,
|
|
"learning_rate": 8.214686587141277e-06,
|
|
"loss": 0.7491,
|
|
"step": 276
|
|
},
|
|
{
|
|
"epoch": 0.3004949488100888,
|
|
"grad_norm": 0.5658873200548137,
|
|
"learning_rate": 8.201194130054342e-06,
|
|
"loss": 0.7508,
|
|
"step": 277
|
|
},
|
|
{
|
|
"epoch": 0.3015797681198725,
|
|
"grad_norm": 0.5669985512277315,
|
|
"learning_rate": 8.187662053522039e-06,
|
|
"loss": 0.7469,
|
|
"step": 278
|
|
},
|
|
{
|
|
"epoch": 0.30266458742965624,
|
|
"grad_norm": 0.3734404207898122,
|
|
"learning_rate": 8.17409052502355e-06,
|
|
"loss": 0.7499,
|
|
"step": 279
|
|
},
|
|
{
|
|
"epoch": 0.30374940673943995,
|
|
"grad_norm": 0.3544398740002715,
|
|
"learning_rate": 8.160479712526326e-06,
|
|
"loss": 0.7492,
|
|
"step": 280
|
|
},
|
|
{
|
|
"epoch": 0.30374940673943995,
|
|
"eval_loss": 0.7486003041267395,
|
|
"eval_runtime": 108.4195,
|
|
"eval_samples_per_second": 87.936,
|
|
"eval_steps_per_second": 0.692,
|
|
"step": 280
|
|
},
|
|
{
|
|
"epoch": 0.30483422604922367,
|
|
"grad_norm": 0.4703481050971955,
|
|
"learning_rate": 8.146829784484024e-06,
|
|
"loss": 0.7499,
|
|
"step": 281
|
|
},
|
|
{
|
|
"epoch": 0.3059190453590074,
|
|
"grad_norm": 0.4413898970855196,
|
|
"learning_rate": 8.133140909834402e-06,
|
|
"loss": 0.7459,
|
|
"step": 282
|
|
},
|
|
{
|
|
"epoch": 0.3070038646687911,
|
|
"grad_norm": 0.39043446148586647,
|
|
"learning_rate": 8.11941325799725e-06,
|
|
"loss": 0.7451,
|
|
"step": 283
|
|
},
|
|
{
|
|
"epoch": 0.3080886839785748,
|
|
"grad_norm": 0.37735097207200924,
|
|
"learning_rate": 8.105646998872275e-06,
|
|
"loss": 0.7501,
|
|
"step": 284
|
|
},
|
|
{
|
|
"epoch": 0.3091735032883585,
|
|
"grad_norm": 0.40138861015101296,
|
|
"learning_rate": 8.091842302837009e-06,
|
|
"loss": 0.7473,
|
|
"step": 285
|
|
},
|
|
{
|
|
"epoch": 0.31025832259814223,
|
|
"grad_norm": 0.4006936450890353,
|
|
"learning_rate": 8.077999340744694e-06,
|
|
"loss": 0.7488,
|
|
"step": 286
|
|
},
|
|
{
|
|
"epoch": 0.31134314190792595,
|
|
"grad_norm": 0.37115953558387643,
|
|
"learning_rate": 8.064118283922173e-06,
|
|
"loss": 0.7484,
|
|
"step": 287
|
|
},
|
|
{
|
|
"epoch": 0.31242796121770966,
|
|
"grad_norm": 0.3578222076322921,
|
|
"learning_rate": 8.050199304167766e-06,
|
|
"loss": 0.7452,
|
|
"step": 288
|
|
},
|
|
{
|
|
"epoch": 0.3135127805274934,
|
|
"grad_norm": 0.3499846039267764,
|
|
"learning_rate": 8.036242573749142e-06,
|
|
"loss": 0.7471,
|
|
"step": 289
|
|
},
|
|
{
|
|
"epoch": 0.3145975998372771,
|
|
"grad_norm": 0.24109238904078859,
|
|
"learning_rate": 8.022248265401196e-06,
|
|
"loss": 0.7475,
|
|
"step": 290
|
|
},
|
|
{
|
|
"epoch": 0.3145975998372771,
|
|
"eval_loss": 0.7470859885215759,
|
|
"eval_runtime": 108.7166,
|
|
"eval_samples_per_second": 87.696,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 290
|
|
},
|
|
{
|
|
"epoch": 0.3156824191470608,
|
|
"grad_norm": 0.27211737689056315,
|
|
"learning_rate": 8.008216552323896e-06,
|
|
"loss": 0.7472,
|
|
"step": 291
|
|
},
|
|
{
|
|
"epoch": 0.3167672384568445,
|
|
"grad_norm": 0.3536030445150533,
|
|
"learning_rate": 7.994147608180153e-06,
|
|
"loss": 0.7483,
|
|
"step": 292
|
|
},
|
|
{
|
|
"epoch": 0.3178520577666282,
|
|
"grad_norm": 0.35791014203746474,
|
|
"learning_rate": 7.98004160709367e-06,
|
|
"loss": 0.7493,
|
|
"step": 293
|
|
},
|
|
{
|
|
"epoch": 0.31893687707641194,
|
|
"grad_norm": 0.25154257811760805,
|
|
"learning_rate": 7.965898723646777e-06,
|
|
"loss": 0.7472,
|
|
"step": 294
|
|
},
|
|
{
|
|
"epoch": 0.32002169638619565,
|
|
"grad_norm": 0.3235703962802231,
|
|
"learning_rate": 7.951719132878279e-06,
|
|
"loss": 0.7465,
|
|
"step": 295
|
|
},
|
|
{
|
|
"epoch": 0.32110651569597937,
|
|
"grad_norm": 0.3736762125868405,
|
|
"learning_rate": 7.93750301028129e-06,
|
|
"loss": 0.7429,
|
|
"step": 296
|
|
},
|
|
{
|
|
"epoch": 0.3221913350057631,
|
|
"grad_norm": 0.3090959154713355,
|
|
"learning_rate": 7.92325053180106e-06,
|
|
"loss": 0.7509,
|
|
"step": 297
|
|
},
|
|
{
|
|
"epoch": 0.3232761543155468,
|
|
"grad_norm": 0.23185495140467952,
|
|
"learning_rate": 7.908961873832788e-06,
|
|
"loss": 0.7476,
|
|
"step": 298
|
|
},
|
|
{
|
|
"epoch": 0.3243609736253305,
|
|
"grad_norm": 0.3156711936339513,
|
|
"learning_rate": 7.894637213219454e-06,
|
|
"loss": 0.7455,
|
|
"step": 299
|
|
},
|
|
{
|
|
"epoch": 0.3254457929351142,
|
|
"grad_norm": 0.3600495814794622,
|
|
"learning_rate": 7.880276727249623e-06,
|
|
"loss": 0.7467,
|
|
"step": 300
|
|
},
|
|
{
|
|
"epoch": 0.3254457929351142,
|
|
"eval_loss": 0.7460697293281555,
|
|
"eval_runtime": 108.7978,
|
|
"eval_samples_per_second": 87.63,
|
|
"eval_steps_per_second": 0.689,
|
|
"step": 300
|
|
},
|
|
{
|
|
"epoch": 0.32653061224489793,
|
|
"grad_norm": 0.30116570450299524,
|
|
"learning_rate": 7.86588059365525e-06,
|
|
"loss": 0.747,
|
|
"step": 301
|
|
},
|
|
{
|
|
"epoch": 0.32761543155468165,
|
|
"grad_norm": 0.2597125478227386,
|
|
"learning_rate": 7.851448990609476e-06,
|
|
"loss": 0.7425,
|
|
"step": 302
|
|
},
|
|
{
|
|
"epoch": 0.32870025086446536,
|
|
"grad_norm": 0.42959501666168687,
|
|
"learning_rate": 7.836982096724438e-06,
|
|
"loss": 0.7482,
|
|
"step": 303
|
|
},
|
|
{
|
|
"epoch": 0.32978507017424913,
|
|
"grad_norm": 0.5062811425805214,
|
|
"learning_rate": 7.82248009104904e-06,
|
|
"loss": 0.7487,
|
|
"step": 304
|
|
},
|
|
{
|
|
"epoch": 0.33086988948403284,
|
|
"grad_norm": 0.31974030710574786,
|
|
"learning_rate": 7.807943153066754e-06,
|
|
"loss": 0.7462,
|
|
"step": 305
|
|
},
|
|
{
|
|
"epoch": 0.33195470879381656,
|
|
"grad_norm": 0.3371842742600445,
|
|
"learning_rate": 7.79337146269338e-06,
|
|
"loss": 0.7448,
|
|
"step": 306
|
|
},
|
|
{
|
|
"epoch": 0.33303952810360027,
|
|
"grad_norm": 0.43175305521955976,
|
|
"learning_rate": 7.77876520027484e-06,
|
|
"loss": 0.7444,
|
|
"step": 307
|
|
},
|
|
{
|
|
"epoch": 0.334124347413384,
|
|
"grad_norm": 0.3917913716525415,
|
|
"learning_rate": 7.764124546584926e-06,
|
|
"loss": 0.7479,
|
|
"step": 308
|
|
},
|
|
{
|
|
"epoch": 0.3352091667231677,
|
|
"grad_norm": 0.3082870201162637,
|
|
"learning_rate": 7.749449682823077e-06,
|
|
"loss": 0.7439,
|
|
"step": 309
|
|
},
|
|
{
|
|
"epoch": 0.3362939860329514,
|
|
"grad_norm": 0.3587933444169138,
|
|
"learning_rate": 7.734740790612137e-06,
|
|
"loss": 0.7419,
|
|
"step": 310
|
|
},
|
|
{
|
|
"epoch": 0.3362939860329514,
|
|
"eval_loss": 0.7452067732810974,
|
|
"eval_runtime": 109.3278,
|
|
"eval_samples_per_second": 87.206,
|
|
"eval_steps_per_second": 0.686,
|
|
"step": 310
|
|
},
|
|
{
|
|
"epoch": 0.3373788053427351,
|
|
"grad_norm": 0.4080653348627163,
|
|
"learning_rate": 7.719998051996087e-06,
|
|
"loss": 0.7439,
|
|
"step": 311
|
|
},
|
|
{
|
|
"epoch": 0.33846362465251884,
|
|
"grad_norm": 0.3349113963655275,
|
|
"learning_rate": 7.705221649437819e-06,
|
|
"loss": 0.7473,
|
|
"step": 312
|
|
},
|
|
{
|
|
"epoch": 0.33954844396230255,
|
|
"grad_norm": 0.3018400509370557,
|
|
"learning_rate": 7.690411765816864e-06,
|
|
"loss": 0.7436,
|
|
"step": 313
|
|
},
|
|
{
|
|
"epoch": 0.34063326327208626,
|
|
"grad_norm": 0.3771840347422595,
|
|
"learning_rate": 7.675568584427125e-06,
|
|
"loss": 0.7481,
|
|
"step": 314
|
|
},
|
|
{
|
|
"epoch": 0.34171808258187,
|
|
"grad_norm": 0.3407511913888794,
|
|
"learning_rate": 7.660692288974618e-06,
|
|
"loss": 0.7418,
|
|
"step": 315
|
|
},
|
|
{
|
|
"epoch": 0.3428029018916537,
|
|
"grad_norm": 0.27293766154428334,
|
|
"learning_rate": 7.64578306357519e-06,
|
|
"loss": 0.7441,
|
|
"step": 316
|
|
},
|
|
{
|
|
"epoch": 0.3438877212014374,
|
|
"grad_norm": 0.3323062643157736,
|
|
"learning_rate": 7.630841092752248e-06,
|
|
"loss": 0.744,
|
|
"step": 317
|
|
},
|
|
{
|
|
"epoch": 0.3449725405112211,
|
|
"grad_norm": 0.358395879207135,
|
|
"learning_rate": 7.615866561434468e-06,
|
|
"loss": 0.7464,
|
|
"step": 318
|
|
},
|
|
{
|
|
"epoch": 0.34605735982100483,
|
|
"grad_norm": 0.30753536422432226,
|
|
"learning_rate": 7.600859654953513e-06,
|
|
"loss": 0.7469,
|
|
"step": 319
|
|
},
|
|
{
|
|
"epoch": 0.34714217913078854,
|
|
"grad_norm": 0.25139155864628004,
|
|
"learning_rate": 7.58582055904173e-06,
|
|
"loss": 0.7453,
|
|
"step": 320
|
|
},
|
|
{
|
|
"epoch": 0.34714217913078854,
|
|
"eval_loss": 0.7439761757850647,
|
|
"eval_runtime": 108.7679,
|
|
"eval_samples_per_second": 87.654,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 320
|
|
},
|
|
{
|
|
"epoch": 0.34822699844057226,
|
|
"grad_norm": 0.27274352956850645,
|
|
"learning_rate": 7.570749459829865e-06,
|
|
"loss": 0.7411,
|
|
"step": 321
|
|
},
|
|
{
|
|
"epoch": 0.34931181775035597,
|
|
"grad_norm": 0.3281154611430217,
|
|
"learning_rate": 7.555646543844747e-06,
|
|
"loss": 0.7445,
|
|
"step": 322
|
|
},
|
|
{
|
|
"epoch": 0.3503966370601397,
|
|
"grad_norm": 0.3578205227364123,
|
|
"learning_rate": 7.540511998006982e-06,
|
|
"loss": 0.7434,
|
|
"step": 323
|
|
},
|
|
{
|
|
"epoch": 0.3514814563699234,
|
|
"grad_norm": 0.33261333446983415,
|
|
"learning_rate": 7.525346009628647e-06,
|
|
"loss": 0.7424,
|
|
"step": 324
|
|
},
|
|
{
|
|
"epoch": 0.3525662756797071,
|
|
"grad_norm": 0.23662027561645424,
|
|
"learning_rate": 7.5101487664109605e-06,
|
|
"loss": 0.7432,
|
|
"step": 325
|
|
},
|
|
{
|
|
"epoch": 0.3536510949894908,
|
|
"grad_norm": 0.2579457554565561,
|
|
"learning_rate": 7.49492045644197e-06,
|
|
"loss": 0.7421,
|
|
"step": 326
|
|
},
|
|
{
|
|
"epoch": 0.35473591429927454,
|
|
"grad_norm": 0.25876584846330475,
|
|
"learning_rate": 7.479661268194217e-06,
|
|
"loss": 0.7434,
|
|
"step": 327
|
|
},
|
|
{
|
|
"epoch": 0.35582073360905825,
|
|
"grad_norm": 0.21029291015757026,
|
|
"learning_rate": 7.4643713905224065e-06,
|
|
"loss": 0.7453,
|
|
"step": 328
|
|
},
|
|
{
|
|
"epoch": 0.35690555291884196,
|
|
"grad_norm": 0.26152934496245606,
|
|
"learning_rate": 7.449051012661073e-06,
|
|
"loss": 0.7423,
|
|
"step": 329
|
|
},
|
|
{
|
|
"epoch": 0.3579903722286257,
|
|
"grad_norm": 0.262606487720371,
|
|
"learning_rate": 7.43370032422223e-06,
|
|
"loss": 0.7413,
|
|
"step": 330
|
|
},
|
|
{
|
|
"epoch": 0.3579903722286257,
|
|
"eval_loss": 0.7429714202880859,
|
|
"eval_runtime": 108.8003,
|
|
"eval_samples_per_second": 87.628,
|
|
"eval_steps_per_second": 0.689,
|
|
"step": 330
|
|
},
|
|
{
|
|
"epoch": 0.3590751915384094,
|
|
"grad_norm": 0.23028217940427023,
|
|
"learning_rate": 7.418319515193032e-06,
|
|
"loss": 0.7423,
|
|
"step": 331
|
|
},
|
|
{
|
|
"epoch": 0.3601600108481931,
|
|
"grad_norm": 0.2960399152265683,
|
|
"learning_rate": 7.402908775933419e-06,
|
|
"loss": 0.7471,
|
|
"step": 332
|
|
},
|
|
{
|
|
"epoch": 0.3612448301579768,
|
|
"grad_norm": 0.31253381303371947,
|
|
"learning_rate": 7.387468297173764e-06,
|
|
"loss": 0.7411,
|
|
"step": 333
|
|
},
|
|
{
|
|
"epoch": 0.36232964946776053,
|
|
"grad_norm": 0.23862488308527482,
|
|
"learning_rate": 7.371998270012504e-06,
|
|
"loss": 0.743,
|
|
"step": 334
|
|
},
|
|
{
|
|
"epoch": 0.36341446877754424,
|
|
"grad_norm": 0.2292933312645703,
|
|
"learning_rate": 7.356498885913784e-06,
|
|
"loss": 0.7479,
|
|
"step": 335
|
|
},
|
|
{
|
|
"epoch": 0.36449928808732796,
|
|
"grad_norm": 0.2754732931383291,
|
|
"learning_rate": 7.340970336705084e-06,
|
|
"loss": 0.7451,
|
|
"step": 336
|
|
},
|
|
{
|
|
"epoch": 0.36558410739711167,
|
|
"grad_norm": 0.22695870049572298,
|
|
"learning_rate": 7.325412814574847e-06,
|
|
"loss": 0.7426,
|
|
"step": 337
|
|
},
|
|
{
|
|
"epoch": 0.3666689267068954,
|
|
"grad_norm": 0.21770343933714315,
|
|
"learning_rate": 7.3098265120700915e-06,
|
|
"loss": 0.7422,
|
|
"step": 338
|
|
},
|
|
{
|
|
"epoch": 0.3677537460166791,
|
|
"grad_norm": 0.23239042788561348,
|
|
"learning_rate": 7.2942116220940406e-06,
|
|
"loss": 0.7414,
|
|
"step": 339
|
|
},
|
|
{
|
|
"epoch": 0.3688385653264628,
|
|
"grad_norm": 0.2035958102762721,
|
|
"learning_rate": 7.278568337903729e-06,
|
|
"loss": 0.7434,
|
|
"step": 340
|
|
},
|
|
{
|
|
"epoch": 0.3688385653264628,
|
|
"eval_loss": 0.7420694231987,
|
|
"eval_runtime": 108.7695,
|
|
"eval_samples_per_second": 87.653,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 340
|
|
},
|
|
{
|
|
"epoch": 0.3699233846362465,
|
|
"grad_norm": 0.25930993206832936,
|
|
"learning_rate": 7.262896853107606e-06,
|
|
"loss": 0.7396,
|
|
"step": 341
|
|
},
|
|
{
|
|
"epoch": 0.37100820394603023,
|
|
"grad_norm": 0.26797052771566304,
|
|
"learning_rate": 7.247197361663152e-06,
|
|
"loss": 0.7424,
|
|
"step": 342
|
|
},
|
|
{
|
|
"epoch": 0.37209302325581395,
|
|
"grad_norm": 0.24610289441173877,
|
|
"learning_rate": 7.2314700578744635e-06,
|
|
"loss": 0.7387,
|
|
"step": 343
|
|
},
|
|
{
|
|
"epoch": 0.37317784256559766,
|
|
"grad_norm": 0.2308231076781789,
|
|
"learning_rate": 7.215715136389862e-06,
|
|
"loss": 0.742,
|
|
"step": 344
|
|
},
|
|
{
|
|
"epoch": 0.3742626618753814,
|
|
"grad_norm": 0.22479038828344666,
|
|
"learning_rate": 7.1999327921994735e-06,
|
|
"loss": 0.7405,
|
|
"step": 345
|
|
},
|
|
{
|
|
"epoch": 0.3753474811851651,
|
|
"grad_norm": 0.25600480485776717,
|
|
"learning_rate": 7.18412322063282e-06,
|
|
"loss": 0.7393,
|
|
"step": 346
|
|
},
|
|
{
|
|
"epoch": 0.3764323004949488,
|
|
"grad_norm": 0.2269217634971134,
|
|
"learning_rate": 7.168286617356406e-06,
|
|
"loss": 0.745,
|
|
"step": 347
|
|
},
|
|
{
|
|
"epoch": 0.3775171198047325,
|
|
"grad_norm": 0.23119750968178973,
|
|
"learning_rate": 7.152423178371286e-06,
|
|
"loss": 0.7425,
|
|
"step": 348
|
|
},
|
|
{
|
|
"epoch": 0.37860193911451623,
|
|
"grad_norm": 0.2243377897178942,
|
|
"learning_rate": 7.136533100010654e-06,
|
|
"loss": 0.7407,
|
|
"step": 349
|
|
},
|
|
{
|
|
"epoch": 0.37968675842429994,
|
|
"grad_norm": 0.22451399833023086,
|
|
"learning_rate": 7.120616578937397e-06,
|
|
"loss": 0.7424,
|
|
"step": 350
|
|
},
|
|
{
|
|
"epoch": 0.37968675842429994,
|
|
"eval_loss": 0.7412689328193665,
|
|
"eval_runtime": 108.4663,
|
|
"eval_samples_per_second": 87.898,
|
|
"eval_steps_per_second": 0.691,
|
|
"step": 350
|
|
},
|
|
{
|
|
"epoch": 0.38077157773408365,
|
|
"grad_norm": 0.2564642617671749,
|
|
"learning_rate": 7.104673812141676e-06,
|
|
"loss": 0.7418,
|
|
"step": 351
|
|
},
|
|
{
|
|
"epoch": 0.38185639704386737,
|
|
"grad_norm": 0.2960848601509433,
|
|
"learning_rate": 7.0887049969384756e-06,
|
|
"loss": 0.7438,
|
|
"step": 352
|
|
},
|
|
{
|
|
"epoch": 0.3829412163536511,
|
|
"grad_norm": 0.23091888442146682,
|
|
"learning_rate": 7.072710330965171e-06,
|
|
"loss": 0.7393,
|
|
"step": 353
|
|
},
|
|
{
|
|
"epoch": 0.3840260356634348,
|
|
"grad_norm": 0.2681419162281407,
|
|
"learning_rate": 7.0566900121790775e-06,
|
|
"loss": 0.7452,
|
|
"step": 354
|
|
},
|
|
{
|
|
"epoch": 0.3851108549732185,
|
|
"grad_norm": 0.22742368238657973,
|
|
"learning_rate": 7.0406442388550016e-06,
|
|
"loss": 0.7401,
|
|
"step": 355
|
|
},
|
|
{
|
|
"epoch": 0.3861956742830022,
|
|
"grad_norm": 0.24788882535022344,
|
|
"learning_rate": 7.024573209582783e-06,
|
|
"loss": 0.7418,
|
|
"step": 356
|
|
},
|
|
{
|
|
"epoch": 0.38728049359278593,
|
|
"grad_norm": 0.25386106926955077,
|
|
"learning_rate": 7.008477123264849e-06,
|
|
"loss": 0.7427,
|
|
"step": 357
|
|
},
|
|
{
|
|
"epoch": 0.38836531290256965,
|
|
"grad_norm": 0.24020641611233573,
|
|
"learning_rate": 6.992356179113735e-06,
|
|
"loss": 0.7432,
|
|
"step": 358
|
|
},
|
|
{
|
|
"epoch": 0.38945013221235336,
|
|
"grad_norm": 0.27832955434324097,
|
|
"learning_rate": 6.9762105766496315e-06,
|
|
"loss": 0.7391,
|
|
"step": 359
|
|
},
|
|
{
|
|
"epoch": 0.3905349515221371,
|
|
"grad_norm": 0.2892777235545787,
|
|
"learning_rate": 6.960040515697918e-06,
|
|
"loss": 0.7443,
|
|
"step": 360
|
|
},
|
|
{
|
|
"epoch": 0.3905349515221371,
|
|
"eval_loss": 0.7404282689094543,
|
|
"eval_runtime": 108.5796,
|
|
"eval_samples_per_second": 87.807,
|
|
"eval_steps_per_second": 0.691,
|
|
"step": 360
|
|
},
|
|
{
|
|
"epoch": 0.3916197708319208,
|
|
"grad_norm": 0.22080220241347404,
|
|
"learning_rate": 6.943846196386673e-06,
|
|
"loss": 0.7413,
|
|
"step": 361
|
|
},
|
|
{
|
|
"epoch": 0.3927045901417045,
|
|
"grad_norm": 0.2761809097565029,
|
|
"learning_rate": 6.927627819144217e-06,
|
|
"loss": 0.7437,
|
|
"step": 362
|
|
},
|
|
{
|
|
"epoch": 0.3937894094514882,
|
|
"grad_norm": 0.29157240828299086,
|
|
"learning_rate": 6.911385584696615e-06,
|
|
"loss": 0.7425,
|
|
"step": 363
|
|
},
|
|
{
|
|
"epoch": 0.39487422876127193,
|
|
"grad_norm": 0.21783923241808995,
|
|
"learning_rate": 6.8951196940652045e-06,
|
|
"loss": 0.7407,
|
|
"step": 364
|
|
},
|
|
{
|
|
"epoch": 0.39595904807105564,
|
|
"grad_norm": 0.25204711843931127,
|
|
"learning_rate": 6.878830348564098e-06,
|
|
"loss": 0.7378,
|
|
"step": 365
|
|
},
|
|
{
|
|
"epoch": 0.39704386738083935,
|
|
"grad_norm": 0.2553266609957623,
|
|
"learning_rate": 6.862517749797703e-06,
|
|
"loss": 0.7416,
|
|
"step": 366
|
|
},
|
|
{
|
|
"epoch": 0.39812868669062307,
|
|
"grad_norm": 0.2600734304620125,
|
|
"learning_rate": 6.846182099658216e-06,
|
|
"loss": 0.7392,
|
|
"step": 367
|
|
},
|
|
{
|
|
"epoch": 0.3992135060004068,
|
|
"grad_norm": 0.2599442716547005,
|
|
"learning_rate": 6.8298236003231264e-06,
|
|
"loss": 0.7414,
|
|
"step": 368
|
|
},
|
|
{
|
|
"epoch": 0.4002983253101905,
|
|
"grad_norm": 0.30360297481374326,
|
|
"learning_rate": 6.8134424542527215e-06,
|
|
"loss": 0.7379,
|
|
"step": 369
|
|
},
|
|
{
|
|
"epoch": 0.4013831446199742,
|
|
"grad_norm": 0.2509784572170823,
|
|
"learning_rate": 6.797038864187564e-06,
|
|
"loss": 0.7388,
|
|
"step": 370
|
|
},
|
|
{
|
|
"epoch": 0.4013831446199742,
|
|
"eval_loss": 0.7396028637886047,
|
|
"eval_runtime": 108.6221,
|
|
"eval_samples_per_second": 87.772,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 370
|
|
},
|
|
{
|
|
"epoch": 0.402467963929758,
|
|
"grad_norm": 0.2068989329070727,
|
|
"learning_rate": 6.780613033146008e-06,
|
|
"loss": 0.7385,
|
|
"step": 371
|
|
},
|
|
{
|
|
"epoch": 0.4035527832395417,
|
|
"grad_norm": 0.2586626388286102,
|
|
"learning_rate": 6.764165164421661e-06,
|
|
"loss": 0.7416,
|
|
"step": 372
|
|
},
|
|
{
|
|
"epoch": 0.4046376025493254,
|
|
"grad_norm": 0.26657867218401476,
|
|
"learning_rate": 6.7476954615808835e-06,
|
|
"loss": 0.7404,
|
|
"step": 373
|
|
},
|
|
{
|
|
"epoch": 0.4057224218591091,
|
|
"grad_norm": 0.24543557206467642,
|
|
"learning_rate": 6.731204128460265e-06,
|
|
"loss": 0.7372,
|
|
"step": 374
|
|
},
|
|
{
|
|
"epoch": 0.40680724116889283,
|
|
"grad_norm": 0.25351491790217756,
|
|
"learning_rate": 6.714691369164099e-06,
|
|
"loss": 0.7394,
|
|
"step": 375
|
|
},
|
|
{
|
|
"epoch": 0.40789206047867654,
|
|
"grad_norm": 0.2805277597400532,
|
|
"learning_rate": 6.6981573880618636e-06,
|
|
"loss": 0.7398,
|
|
"step": 376
|
|
},
|
|
{
|
|
"epoch": 0.40897687978846026,
|
|
"grad_norm": 0.29838785276264307,
|
|
"learning_rate": 6.681602389785683e-06,
|
|
"loss": 0.7423,
|
|
"step": 377
|
|
},
|
|
{
|
|
"epoch": 0.41006169909824397,
|
|
"grad_norm": 0.2596895547792377,
|
|
"learning_rate": 6.665026579227802e-06,
|
|
"loss": 0.7386,
|
|
"step": 378
|
|
},
|
|
{
|
|
"epoch": 0.4111465184080277,
|
|
"grad_norm": 0.23646818360935362,
|
|
"learning_rate": 6.648430161538047e-06,
|
|
"loss": 0.7382,
|
|
"step": 379
|
|
},
|
|
{
|
|
"epoch": 0.4122313377178114,
|
|
"grad_norm": 0.2437039188612826,
|
|
"learning_rate": 6.631813342121289e-06,
|
|
"loss": 0.7383,
|
|
"step": 380
|
|
},
|
|
{
|
|
"epoch": 0.4122313377178114,
|
|
"eval_loss": 0.7389257550239563,
|
|
"eval_runtime": 108.7067,
|
|
"eval_samples_per_second": 87.704,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 380
|
|
},
|
|
{
|
|
"epoch": 0.4133161570275951,
|
|
"grad_norm": 0.21812226663907686,
|
|
"learning_rate": 6.6151763266348975e-06,
|
|
"loss": 0.7398,
|
|
"step": 381
|
|
},
|
|
{
|
|
"epoch": 0.4144009763373788,
|
|
"grad_norm": 0.21982244825408284,
|
|
"learning_rate": 6.598519320986201e-06,
|
|
"loss": 0.7386,
|
|
"step": 382
|
|
},
|
|
{
|
|
"epoch": 0.41548579564716254,
|
|
"grad_norm": 0.2449765304599977,
|
|
"learning_rate": 6.5818425313299325e-06,
|
|
"loss": 0.7392,
|
|
"step": 383
|
|
},
|
|
{
|
|
"epoch": 0.41657061495694625,
|
|
"grad_norm": 0.2844945982939222,
|
|
"learning_rate": 6.5651461640656825e-06,
|
|
"loss": 0.7391,
|
|
"step": 384
|
|
},
|
|
{
|
|
"epoch": 0.41765543426672996,
|
|
"grad_norm": 0.23494136956442008,
|
|
"learning_rate": 6.5484304258353435e-06,
|
|
"loss": 0.7378,
|
|
"step": 385
|
|
},
|
|
{
|
|
"epoch": 0.4187402535765137,
|
|
"grad_norm": 0.24562274123009595,
|
|
"learning_rate": 6.53169552352055e-06,
|
|
"loss": 0.7429,
|
|
"step": 386
|
|
},
|
|
{
|
|
"epoch": 0.4198250728862974,
|
|
"grad_norm": 0.26227200036290854,
|
|
"learning_rate": 6.514941664240122e-06,
|
|
"loss": 0.7396,
|
|
"step": 387
|
|
},
|
|
{
|
|
"epoch": 0.4209098921960811,
|
|
"grad_norm": 0.2615429246835793,
|
|
"learning_rate": 6.498169055347498e-06,
|
|
"loss": 0.7345,
|
|
"step": 388
|
|
},
|
|
{
|
|
"epoch": 0.4219947115058648,
|
|
"grad_norm": 0.316237656594935,
|
|
"learning_rate": 6.481377904428171e-06,
|
|
"loss": 0.7342,
|
|
"step": 389
|
|
},
|
|
{
|
|
"epoch": 0.42307953081564853,
|
|
"grad_norm": 0.27503244758564993,
|
|
"learning_rate": 6.4645684192971195e-06,
|
|
"loss": 0.7375,
|
|
"step": 390
|
|
},
|
|
{
|
|
"epoch": 0.42307953081564853,
|
|
"eval_loss": 0.7380961179733276,
|
|
"eval_runtime": 108.5728,
|
|
"eval_samples_per_second": 87.812,
|
|
"eval_steps_per_second": 0.691,
|
|
"step": 390
|
|
},
|
|
{
|
|
"epoch": 0.42416435012543224,
|
|
"grad_norm": 0.2100505409018419,
|
|
"learning_rate": 6.447740807996232e-06,
|
|
"loss": 0.7403,
|
|
"step": 391
|
|
},
|
|
{
|
|
"epoch": 0.42524916943521596,
|
|
"grad_norm": 0.2346973440075669,
|
|
"learning_rate": 6.430895278791739e-06,
|
|
"loss": 0.7388,
|
|
"step": 392
|
|
},
|
|
{
|
|
"epoch": 0.42633398874499967,
|
|
"grad_norm": 0.32481498647324375,
|
|
"learning_rate": 6.414032040171627e-06,
|
|
"loss": 0.7383,
|
|
"step": 393
|
|
},
|
|
{
|
|
"epoch": 0.4274188080547834,
|
|
"grad_norm": 0.27715402097878344,
|
|
"learning_rate": 6.397151300843065e-06,
|
|
"loss": 0.7401,
|
|
"step": 394
|
|
},
|
|
{
|
|
"epoch": 0.4285036273645671,
|
|
"grad_norm": 0.22340245769952607,
|
|
"learning_rate": 6.380253269729816e-06,
|
|
"loss": 0.7391,
|
|
"step": 395
|
|
},
|
|
{
|
|
"epoch": 0.4295884466743508,
|
|
"grad_norm": 0.26387297611191257,
|
|
"learning_rate": 6.363338155969658e-06,
|
|
"loss": 0.7367,
|
|
"step": 396
|
|
},
|
|
{
|
|
"epoch": 0.4306732659841345,
|
|
"grad_norm": 0.2704074945545565,
|
|
"learning_rate": 6.346406168911787e-06,
|
|
"loss": 0.7385,
|
|
"step": 397
|
|
},
|
|
{
|
|
"epoch": 0.43175808529391824,
|
|
"grad_norm": 0.22282484781205528,
|
|
"learning_rate": 6.329457518114237e-06,
|
|
"loss": 0.7413,
|
|
"step": 398
|
|
},
|
|
{
|
|
"epoch": 0.43284290460370195,
|
|
"grad_norm": 0.2562608743683754,
|
|
"learning_rate": 6.312492413341274e-06,
|
|
"loss": 0.7402,
|
|
"step": 399
|
|
},
|
|
{
|
|
"epoch": 0.43392772391348566,
|
|
"grad_norm": 0.21893354922393973,
|
|
"learning_rate": 6.295511064560808e-06,
|
|
"loss": 0.736,
|
|
"step": 400
|
|
},
|
|
{
|
|
"epoch": 0.43392772391348566,
|
|
"eval_loss": 0.7374843955039978,
|
|
"eval_runtime": 109.4223,
|
|
"eval_samples_per_second": 87.13,
|
|
"eval_steps_per_second": 0.685,
|
|
"step": 400
|
|
},
|
|
{
|
|
"epoch": 0.4350125432232694,
|
|
"grad_norm": 0.24345200142561457,
|
|
"learning_rate": 6.278513681941793e-06,
|
|
"loss": 0.7361,
|
|
"step": 401
|
|
},
|
|
{
|
|
"epoch": 0.4360973625330531,
|
|
"grad_norm": 0.24253672616097802,
|
|
"learning_rate": 6.261500475851625e-06,
|
|
"loss": 0.737,
|
|
"step": 402
|
|
},
|
|
{
|
|
"epoch": 0.4371821818428368,
|
|
"grad_norm": 0.2376999226855751,
|
|
"learning_rate": 6.244471656853538e-06,
|
|
"loss": 0.7377,
|
|
"step": 403
|
|
},
|
|
{
|
|
"epoch": 0.4382670011526205,
|
|
"grad_norm": 0.2354206750040896,
|
|
"learning_rate": 6.227427435703997e-06,
|
|
"loss": 0.739,
|
|
"step": 404
|
|
},
|
|
{
|
|
"epoch": 0.43935182046240423,
|
|
"grad_norm": 0.3051544078019008,
|
|
"learning_rate": 6.210368023350094e-06,
|
|
"loss": 0.7374,
|
|
"step": 405
|
|
},
|
|
{
|
|
"epoch": 0.44043663977218794,
|
|
"grad_norm": 0.25896047178178727,
|
|
"learning_rate": 6.193293630926933e-06,
|
|
"loss": 0.7379,
|
|
"step": 406
|
|
},
|
|
{
|
|
"epoch": 0.44152145908197166,
|
|
"grad_norm": 0.21060632035092675,
|
|
"learning_rate": 6.176204469755021e-06,
|
|
"loss": 0.7358,
|
|
"step": 407
|
|
},
|
|
{
|
|
"epoch": 0.44260627839175537,
|
|
"grad_norm": 0.265287637808988,
|
|
"learning_rate": 6.1591007513376425e-06,
|
|
"loss": 0.7345,
|
|
"step": 408
|
|
},
|
|
{
|
|
"epoch": 0.4436910977015391,
|
|
"grad_norm": 0.24986223470603142,
|
|
"learning_rate": 6.141982687358255e-06,
|
|
"loss": 0.7356,
|
|
"step": 409
|
|
},
|
|
{
|
|
"epoch": 0.4447759170113228,
|
|
"grad_norm": 0.24992318775486172,
|
|
"learning_rate": 6.124850489677865e-06,
|
|
"loss": 0.7388,
|
|
"step": 410
|
|
},
|
|
{
|
|
"epoch": 0.4447759170113228,
|
|
"eval_loss": 0.7367204427719116,
|
|
"eval_runtime": 108.695,
|
|
"eval_samples_per_second": 87.713,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 410
|
|
},
|
|
{
|
|
"epoch": 0.4458607363211065,
|
|
"grad_norm": 0.23125436571593908,
|
|
"learning_rate": 6.1077043703323964e-06,
|
|
"loss": 0.7362,
|
|
"step": 411
|
|
},
|
|
{
|
|
"epoch": 0.4469455556308902,
|
|
"grad_norm": 0.2195206266766565,
|
|
"learning_rate": 6.0905445415300835e-06,
|
|
"loss": 0.7358,
|
|
"step": 412
|
|
},
|
|
{
|
|
"epoch": 0.44803037494067394,
|
|
"grad_norm": 0.22978816149553494,
|
|
"learning_rate": 6.073371215648824e-06,
|
|
"loss": 0.7341,
|
|
"step": 413
|
|
},
|
|
{
|
|
"epoch": 0.44911519425045765,
|
|
"grad_norm": 0.2044126050489451,
|
|
"learning_rate": 6.056184605233576e-06,
|
|
"loss": 0.7397,
|
|
"step": 414
|
|
},
|
|
{
|
|
"epoch": 0.45020001356024136,
|
|
"grad_norm": 0.2431328228598919,
|
|
"learning_rate": 6.0389849229936995e-06,
|
|
"loss": 0.7324,
|
|
"step": 415
|
|
},
|
|
{
|
|
"epoch": 0.4512848328700251,
|
|
"grad_norm": 0.2124680032728403,
|
|
"learning_rate": 6.021772381800344e-06,
|
|
"loss": 0.7388,
|
|
"step": 416
|
|
},
|
|
{
|
|
"epoch": 0.4523696521798088,
|
|
"grad_norm": 0.22907518203097438,
|
|
"learning_rate": 6.004547194683806e-06,
|
|
"loss": 0.7382,
|
|
"step": 417
|
|
},
|
|
{
|
|
"epoch": 0.4534544714895925,
|
|
"grad_norm": 0.21543245283430915,
|
|
"learning_rate": 5.987309574830897e-06,
|
|
"loss": 0.7378,
|
|
"step": 418
|
|
},
|
|
{
|
|
"epoch": 0.4545392907993762,
|
|
"grad_norm": 0.19445002534528355,
|
|
"learning_rate": 5.970059735582295e-06,
|
|
"loss": 0.7352,
|
|
"step": 419
|
|
},
|
|
{
|
|
"epoch": 0.45562411010915993,
|
|
"grad_norm": 0.20092408059131545,
|
|
"learning_rate": 5.9527978904299156e-06,
|
|
"loss": 0.7325,
|
|
"step": 420
|
|
},
|
|
{
|
|
"epoch": 0.45562411010915993,
|
|
"eval_loss": 0.7360592484474182,
|
|
"eval_runtime": 108.601,
|
|
"eval_samples_per_second": 87.789,
|
|
"eval_steps_per_second": 0.691,
|
|
"step": 420
|
|
},
|
|
{
|
|
"epoch": 0.45670892941894364,
|
|
"grad_norm": 0.20942902891105614,
|
|
"learning_rate": 5.935524253014263e-06,
|
|
"loss": 0.7345,
|
|
"step": 421
|
|
},
|
|
{
|
|
"epoch": 0.45779374872872736,
|
|
"grad_norm": 0.21326648446138302,
|
|
"learning_rate": 5.918239037121791e-06,
|
|
"loss": 0.733,
|
|
"step": 422
|
|
},
|
|
{
|
|
"epoch": 0.45887856803851107,
|
|
"grad_norm": 0.19801418492230252,
|
|
"learning_rate": 5.9009424566822515e-06,
|
|
"loss": 0.7377,
|
|
"step": 423
|
|
},
|
|
{
|
|
"epoch": 0.4599633873482948,
|
|
"grad_norm": 0.22467857965205296,
|
|
"learning_rate": 5.8836347257660485e-06,
|
|
"loss": 0.7357,
|
|
"step": 424
|
|
},
|
|
{
|
|
"epoch": 0.4610482066580785,
|
|
"grad_norm": 0.1942099576377276,
|
|
"learning_rate": 5.86631605858159e-06,
|
|
"loss": 0.7359,
|
|
"step": 425
|
|
},
|
|
{
|
|
"epoch": 0.4621330259678622,
|
|
"grad_norm": 0.22370006156205846,
|
|
"learning_rate": 5.848986669472637e-06,
|
|
"loss": 0.7385,
|
|
"step": 426
|
|
},
|
|
{
|
|
"epoch": 0.4632178452776459,
|
|
"grad_norm": 0.23461327887740901,
|
|
"learning_rate": 5.831646772915651e-06,
|
|
"loss": 0.734,
|
|
"step": 427
|
|
},
|
|
{
|
|
"epoch": 0.46430266458742964,
|
|
"grad_norm": 0.23034977016579453,
|
|
"learning_rate": 5.814296583517135e-06,
|
|
"loss": 0.7344,
|
|
"step": 428
|
|
},
|
|
{
|
|
"epoch": 0.46538748389721335,
|
|
"grad_norm": 0.21750438436915417,
|
|
"learning_rate": 5.796936316010984e-06,
|
|
"loss": 0.7367,
|
|
"step": 429
|
|
},
|
|
{
|
|
"epoch": 0.46647230320699706,
|
|
"grad_norm": 0.22841954936308745,
|
|
"learning_rate": 5.779566185255823e-06,
|
|
"loss": 0.7335,
|
|
"step": 430
|
|
},
|
|
{
|
|
"epoch": 0.46647230320699706,
|
|
"eval_loss": 0.7354937195777893,
|
|
"eval_runtime": 108.3841,
|
|
"eval_samples_per_second": 87.965,
|
|
"eval_steps_per_second": 0.692,
|
|
"step": 430
|
|
},
|
|
{
|
|
"epoch": 0.4675571225167808,
|
|
"grad_norm": 0.2502062087756078,
|
|
"learning_rate": 5.7621864062323484e-06,
|
|
"loss": 0.7364,
|
|
"step": 431
|
|
},
|
|
{
|
|
"epoch": 0.4686419418265645,
|
|
"grad_norm": 0.2102212571314435,
|
|
"learning_rate": 5.744797194040672e-06,
|
|
"loss": 0.7334,
|
|
"step": 432
|
|
},
|
|
{
|
|
"epoch": 0.4697267611363482,
|
|
"grad_norm": 0.20734252560627506,
|
|
"learning_rate": 5.727398763897648e-06,
|
|
"loss": 0.7355,
|
|
"step": 433
|
|
},
|
|
{
|
|
"epoch": 0.4708115804461319,
|
|
"grad_norm": 0.18853441875991667,
|
|
"learning_rate": 5.709991331134224e-06,
|
|
"loss": 0.7363,
|
|
"step": 434
|
|
},
|
|
{
|
|
"epoch": 0.47189639975591563,
|
|
"grad_norm": 0.2270702539080567,
|
|
"learning_rate": 5.69257511119276e-06,
|
|
"loss": 0.7357,
|
|
"step": 435
|
|
},
|
|
{
|
|
"epoch": 0.47298121906569934,
|
|
"grad_norm": 0.1970447518104085,
|
|
"learning_rate": 5.675150319624375e-06,
|
|
"loss": 0.7344,
|
|
"step": 436
|
|
},
|
|
{
|
|
"epoch": 0.47406603837548306,
|
|
"grad_norm": 0.23973739416294565,
|
|
"learning_rate": 5.657717172086278e-06,
|
|
"loss": 0.7364,
|
|
"step": 437
|
|
},
|
|
{
|
|
"epoch": 0.4751508576852668,
|
|
"grad_norm": 0.28314313613476666,
|
|
"learning_rate": 5.6402758843390844e-06,
|
|
"loss": 0.7383,
|
|
"step": 438
|
|
},
|
|
{
|
|
"epoch": 0.47623567699505054,
|
|
"grad_norm": 0.1916371805667599,
|
|
"learning_rate": 5.622826672244169e-06,
|
|
"loss": 0.734,
|
|
"step": 439
|
|
},
|
|
{
|
|
"epoch": 0.47732049630483425,
|
|
"grad_norm": 0.2354299932558386,
|
|
"learning_rate": 5.6053697517609725e-06,
|
|
"loss": 0.7347,
|
|
"step": 440
|
|
},
|
|
{
|
|
"epoch": 0.47732049630483425,
|
|
"eval_loss": 0.7349258661270142,
|
|
"eval_runtime": 108.6897,
|
|
"eval_samples_per_second": 87.718,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 440
|
|
},
|
|
{
|
|
"epoch": 0.47840531561461797,
|
|
"grad_norm": 0.2803627977735078,
|
|
"learning_rate": 5.5879053389443435e-06,
|
|
"loss": 0.7374,
|
|
"step": 441
|
|
},
|
|
{
|
|
"epoch": 0.4794901349244017,
|
|
"grad_norm": 0.2537444696570077,
|
|
"learning_rate": 5.570433649941855e-06,
|
|
"loss": 0.7378,
|
|
"step": 442
|
|
},
|
|
{
|
|
"epoch": 0.4805749542341854,
|
|
"grad_norm": 0.23908816966210633,
|
|
"learning_rate": 5.552954900991139e-06,
|
|
"loss": 0.734,
|
|
"step": 443
|
|
},
|
|
{
|
|
"epoch": 0.4816597735439691,
|
|
"grad_norm": 0.22445374681545627,
|
|
"learning_rate": 5.535469308417198e-06,
|
|
"loss": 0.7348,
|
|
"step": 444
|
|
},
|
|
{
|
|
"epoch": 0.4827445928537528,
|
|
"grad_norm": 0.2539692098081985,
|
|
"learning_rate": 5.5179770886297405e-06,
|
|
"loss": 0.7359,
|
|
"step": 445
|
|
},
|
|
{
|
|
"epoch": 0.48382941216353653,
|
|
"grad_norm": 0.2448220066882044,
|
|
"learning_rate": 5.500478458120493e-06,
|
|
"loss": 0.7362,
|
|
"step": 446
|
|
},
|
|
{
|
|
"epoch": 0.48491423147332025,
|
|
"grad_norm": 0.25630523224056756,
|
|
"learning_rate": 5.482973633460524e-06,
|
|
"loss": 0.7365,
|
|
"step": 447
|
|
},
|
|
{
|
|
"epoch": 0.48599905078310396,
|
|
"grad_norm": 0.22610992337857022,
|
|
"learning_rate": 5.465462831297564e-06,
|
|
"loss": 0.7343,
|
|
"step": 448
|
|
},
|
|
{
|
|
"epoch": 0.48708387009288767,
|
|
"grad_norm": 0.30219127502735,
|
|
"learning_rate": 5.447946268353324e-06,
|
|
"loss": 0.7352,
|
|
"step": 449
|
|
},
|
|
{
|
|
"epoch": 0.4881686894026714,
|
|
"grad_norm": 0.2517123588413187,
|
|
"learning_rate": 5.430424161420817e-06,
|
|
"loss": 0.7357,
|
|
"step": 450
|
|
},
|
|
{
|
|
"epoch": 0.4881686894026714,
|
|
"eval_loss": 0.7343276143074036,
|
|
"eval_runtime": 108.3618,
|
|
"eval_samples_per_second": 87.983,
|
|
"eval_steps_per_second": 0.692,
|
|
"step": 450
|
|
},
|
|
{
|
|
"epoch": 0.4892535087124551,
|
|
"grad_norm": 0.24872326056896576,
|
|
"learning_rate": 5.412896727361663e-06,
|
|
"loss": 0.7322,
|
|
"step": 451
|
|
},
|
|
{
|
|
"epoch": 0.4903383280222388,
|
|
"grad_norm": 0.2530744795907447,
|
|
"learning_rate": 5.395364183103418e-06,
|
|
"loss": 0.7322,
|
|
"step": 452
|
|
},
|
|
{
|
|
"epoch": 0.4914231473320225,
|
|
"grad_norm": 0.2701453199725646,
|
|
"learning_rate": 5.3778267456368836e-06,
|
|
"loss": 0.7336,
|
|
"step": 453
|
|
},
|
|
{
|
|
"epoch": 0.49250796664180624,
|
|
"grad_norm": 0.25539652632182863,
|
|
"learning_rate": 5.3602846320134216e-06,
|
|
"loss": 0.7323,
|
|
"step": 454
|
|
},
|
|
{
|
|
"epoch": 0.49359278595158995,
|
|
"grad_norm": 0.20991044834587508,
|
|
"learning_rate": 5.34273805934227e-06,
|
|
"loss": 0.7348,
|
|
"step": 455
|
|
},
|
|
{
|
|
"epoch": 0.49467760526137367,
|
|
"grad_norm": 0.2867980936996136,
|
|
"learning_rate": 5.325187244787848e-06,
|
|
"loss": 0.7338,
|
|
"step": 456
|
|
},
|
|
{
|
|
"epoch": 0.4957624245711574,
|
|
"grad_norm": 0.2422259007921334,
|
|
"learning_rate": 5.307632405567084e-06,
|
|
"loss": 0.7347,
|
|
"step": 457
|
|
},
|
|
{
|
|
"epoch": 0.4968472438809411,
|
|
"grad_norm": 0.2679236313890693,
|
|
"learning_rate": 5.290073758946705e-06,
|
|
"loss": 0.7339,
|
|
"step": 458
|
|
},
|
|
{
|
|
"epoch": 0.4979320631907248,
|
|
"grad_norm": 0.21091521318926196,
|
|
"learning_rate": 5.272511522240574e-06,
|
|
"loss": 0.7344,
|
|
"step": 459
|
|
},
|
|
{
|
|
"epoch": 0.4990168825005085,
|
|
"grad_norm": 0.27650703535788657,
|
|
"learning_rate": 5.254945912806977e-06,
|
|
"loss": 0.7323,
|
|
"step": 460
|
|
},
|
|
{
|
|
"epoch": 0.4990168825005085,
|
|
"eval_loss": 0.733704149723053,
|
|
"eval_runtime": 108.393,
|
|
"eval_samples_per_second": 87.958,
|
|
"eval_steps_per_second": 0.692,
|
|
"step": 460
|
|
},
|
|
{
|
|
"epoch": 0.5001017018102922,
|
|
"grad_norm": 0.227218346086516,
|
|
"learning_rate": 5.237377148045942e-06,
|
|
"loss": 0.7354,
|
|
"step": 461
|
|
},
|
|
{
|
|
"epoch": 0.5011865211200759,
|
|
"grad_norm": 0.25506532184573566,
|
|
"learning_rate": 5.219805445396558e-06,
|
|
"loss": 0.7352,
|
|
"step": 462
|
|
},
|
|
{
|
|
"epoch": 0.5022713404298597,
|
|
"grad_norm": 0.20879772361101673,
|
|
"learning_rate": 5.202231022334262e-06,
|
|
"loss": 0.729,
|
|
"step": 463
|
|
},
|
|
{
|
|
"epoch": 0.5033561597396433,
|
|
"grad_norm": 0.24241179715472216,
|
|
"learning_rate": 5.184654096368172e-06,
|
|
"loss": 0.7346,
|
|
"step": 464
|
|
},
|
|
{
|
|
"epoch": 0.5044409790494271,
|
|
"grad_norm": 0.18914490208247275,
|
|
"learning_rate": 5.1670748850383734e-06,
|
|
"loss": 0.7328,
|
|
"step": 465
|
|
},
|
|
{
|
|
"epoch": 0.5055257983592107,
|
|
"grad_norm": 0.2682231520390539,
|
|
"learning_rate": 5.149493605913244e-06,
|
|
"loss": 0.7345,
|
|
"step": 466
|
|
},
|
|
{
|
|
"epoch": 0.5066106176689945,
|
|
"grad_norm": 0.2328731862075947,
|
|
"learning_rate": 5.131910476586747e-06,
|
|
"loss": 0.7359,
|
|
"step": 467
|
|
},
|
|
{
|
|
"epoch": 0.5076954369787782,
|
|
"grad_norm": 0.24741560042413938,
|
|
"learning_rate": 5.1143257146757495e-06,
|
|
"loss": 0.7345,
|
|
"step": 468
|
|
},
|
|
{
|
|
"epoch": 0.5087802562885619,
|
|
"grad_norm": 0.20835863980252986,
|
|
"learning_rate": 5.096739537817324e-06,
|
|
"loss": 0.7325,
|
|
"step": 469
|
|
},
|
|
{
|
|
"epoch": 0.5098650755983456,
|
|
"grad_norm": 0.22220444993198457,
|
|
"learning_rate": 5.07915216366605e-06,
|
|
"loss": 0.7334,
|
|
"step": 470
|
|
},
|
|
{
|
|
"epoch": 0.5098650755983456,
|
|
"eval_loss": 0.7331451773643494,
|
|
"eval_runtime": 108.2951,
|
|
"eval_samples_per_second": 88.037,
|
|
"eval_steps_per_second": 0.693,
|
|
"step": 470
|
|
},
|
|
{
|
|
"epoch": 0.5109498949081294,
|
|
"grad_norm": 0.19346540419369498,
|
|
"learning_rate": 5.061563809891331e-06,
|
|
"loss": 0.7348,
|
|
"step": 471
|
|
},
|
|
{
|
|
"epoch": 0.5120347142179131,
|
|
"grad_norm": 0.20956923031444313,
|
|
"learning_rate": 5.0439746941746914e-06,
|
|
"loss": 0.7327,
|
|
"step": 472
|
|
},
|
|
{
|
|
"epoch": 0.5131195335276968,
|
|
"grad_norm": 0.1948067717784309,
|
|
"learning_rate": 5.026385034207087e-06,
|
|
"loss": 0.7334,
|
|
"step": 473
|
|
},
|
|
{
|
|
"epoch": 0.5142043528374806,
|
|
"grad_norm": 0.2078607900747825,
|
|
"learning_rate": 5.00879504768621e-06,
|
|
"loss": 0.7322,
|
|
"step": 474
|
|
},
|
|
{
|
|
"epoch": 0.5152891721472642,
|
|
"grad_norm": 0.20015167654418073,
|
|
"learning_rate": 4.991204952313793e-06,
|
|
"loss": 0.7346,
|
|
"step": 475
|
|
},
|
|
{
|
|
"epoch": 0.516373991457048,
|
|
"grad_norm": 0.19111860028800567,
|
|
"learning_rate": 4.9736149657929136e-06,
|
|
"loss": 0.7359,
|
|
"step": 476
|
|
},
|
|
{
|
|
"epoch": 0.5174588107668316,
|
|
"grad_norm": 0.21180553096590882,
|
|
"learning_rate": 4.956025305825311e-06,
|
|
"loss": 0.7328,
|
|
"step": 477
|
|
},
|
|
{
|
|
"epoch": 0.5185436300766154,
|
|
"grad_norm": 0.19381515008692068,
|
|
"learning_rate": 4.93843619010867e-06,
|
|
"loss": 0.7333,
|
|
"step": 478
|
|
},
|
|
{
|
|
"epoch": 0.5196284493863991,
|
|
"grad_norm": 0.18719529016986644,
|
|
"learning_rate": 4.92084783633395e-06,
|
|
"loss": 0.7329,
|
|
"step": 479
|
|
},
|
|
{
|
|
"epoch": 0.5207132686961828,
|
|
"grad_norm": 0.20144464644999407,
|
|
"learning_rate": 4.903260462182679e-06,
|
|
"loss": 0.7323,
|
|
"step": 480
|
|
},
|
|
{
|
|
"epoch": 0.5207132686961828,
|
|
"eval_loss": 0.732681393623352,
|
|
"eval_runtime": 108.7417,
|
|
"eval_samples_per_second": 87.676,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 480
|
|
},
|
|
{
|
|
"epoch": 0.5217980880059665,
|
|
"grad_norm": 0.22355889813141624,
|
|
"learning_rate": 4.8856742853242504e-06,
|
|
"loss": 0.7299,
|
|
"step": 481
|
|
},
|
|
{
|
|
"epoch": 0.5228829073157503,
|
|
"grad_norm": 0.18565478155173432,
|
|
"learning_rate": 4.868089523413255e-06,
|
|
"loss": 0.726,
|
|
"step": 482
|
|
},
|
|
{
|
|
"epoch": 0.5239677266255339,
|
|
"grad_norm": 0.21221933567144674,
|
|
"learning_rate": 4.850506394086758e-06,
|
|
"loss": 0.7289,
|
|
"step": 483
|
|
},
|
|
{
|
|
"epoch": 0.5250525459353177,
|
|
"grad_norm": 0.2034760004016191,
|
|
"learning_rate": 4.832925114961629e-06,
|
|
"loss": 0.7341,
|
|
"step": 484
|
|
},
|
|
{
|
|
"epoch": 0.5261373652451014,
|
|
"grad_norm": 0.19378608630400168,
|
|
"learning_rate": 4.8153459036318295e-06,
|
|
"loss": 0.7332,
|
|
"step": 485
|
|
},
|
|
{
|
|
"epoch": 0.5272221845548851,
|
|
"grad_norm": 0.1896712553317738,
|
|
"learning_rate": 4.79776897766574e-06,
|
|
"loss": 0.7325,
|
|
"step": 486
|
|
},
|
|
{
|
|
"epoch": 0.5283070038646688,
|
|
"grad_norm": 0.20014354653616181,
|
|
"learning_rate": 4.780194554603444e-06,
|
|
"loss": 0.73,
|
|
"step": 487
|
|
},
|
|
{
|
|
"epoch": 0.5293918231744525,
|
|
"grad_norm": 0.20443916179047886,
|
|
"learning_rate": 4.762622851954058e-06,
|
|
"loss": 0.7336,
|
|
"step": 488
|
|
},
|
|
{
|
|
"epoch": 0.5304766424842362,
|
|
"grad_norm": 0.20239934234036272,
|
|
"learning_rate": 4.745054087193025e-06,
|
|
"loss": 0.7339,
|
|
"step": 489
|
|
},
|
|
{
|
|
"epoch": 0.53156146179402,
|
|
"grad_norm": 0.20051945304019603,
|
|
"learning_rate": 4.7274884777594265e-06,
|
|
"loss": 0.7334,
|
|
"step": 490
|
|
},
|
|
{
|
|
"epoch": 0.53156146179402,
|
|
"eval_loss": 0.732123613357544,
|
|
"eval_runtime": 108.9346,
|
|
"eval_samples_per_second": 87.52,
|
|
"eval_steps_per_second": 0.688,
|
|
"step": 490
|
|
},
|
|
{
|
|
"epoch": 0.5326462811038036,
|
|
"grad_norm": 0.19350589563544005,
|
|
"learning_rate": 4.709926241053296e-06,
|
|
"loss": 0.7356,
|
|
"step": 491
|
|
},
|
|
{
|
|
"epoch": 0.5337311004135874,
|
|
"grad_norm": 0.2056225401685897,
|
|
"learning_rate": 4.692367594432919e-06,
|
|
"loss": 0.7333,
|
|
"step": 492
|
|
},
|
|
{
|
|
"epoch": 0.5348159197233711,
|
|
"grad_norm": 0.21513896967475132,
|
|
"learning_rate": 4.674812755212154e-06,
|
|
"loss": 0.7332,
|
|
"step": 493
|
|
},
|
|
{
|
|
"epoch": 0.5359007390331548,
|
|
"grad_norm": 0.1902067755419533,
|
|
"learning_rate": 4.657261940657732e-06,
|
|
"loss": 0.7302,
|
|
"step": 494
|
|
},
|
|
{
|
|
"epoch": 0.5369855583429385,
|
|
"grad_norm": 0.21402370904640888,
|
|
"learning_rate": 4.639715367986578e-06,
|
|
"loss": 0.732,
|
|
"step": 495
|
|
},
|
|
{
|
|
"epoch": 0.5380703776527223,
|
|
"grad_norm": 0.19806533610516014,
|
|
"learning_rate": 4.622173254363117e-06,
|
|
"loss": 0.7321,
|
|
"step": 496
|
|
},
|
|
{
|
|
"epoch": 0.5391551969625059,
|
|
"grad_norm": 0.193145517897021,
|
|
"learning_rate": 4.604635816896583e-06,
|
|
"loss": 0.7323,
|
|
"step": 497
|
|
},
|
|
{
|
|
"epoch": 0.5402400162722897,
|
|
"grad_norm": 0.18969830943803712,
|
|
"learning_rate": 4.587103272638339e-06,
|
|
"loss": 0.7302,
|
|
"step": 498
|
|
},
|
|
{
|
|
"epoch": 0.5413248355820733,
|
|
"grad_norm": 0.21037790338680454,
|
|
"learning_rate": 4.569575838579184e-06,
|
|
"loss": 0.7311,
|
|
"step": 499
|
|
},
|
|
{
|
|
"epoch": 0.5424096548918571,
|
|
"grad_norm": 0.18488892844134952,
|
|
"learning_rate": 4.5520537316466775e-06,
|
|
"loss": 0.7328,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5424096548918571,
|
|
"eval_loss": 0.7317008972167969,
|
|
"eval_runtime": 109.3016,
|
|
"eval_samples_per_second": 87.227,
|
|
"eval_steps_per_second": 0.686,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5434944742016408,
|
|
"grad_norm": 0.2185697237459928,
|
|
"learning_rate": 4.534537168702437e-06,
|
|
"loss": 0.7317,
|
|
"step": 501
|
|
},
|
|
{
|
|
"epoch": 0.5445792935114245,
|
|
"grad_norm": 0.21224465792805108,
|
|
"learning_rate": 4.517026366539477e-06,
|
|
"loss": 0.7319,
|
|
"step": 502
|
|
},
|
|
{
|
|
"epoch": 0.5456641128212082,
|
|
"grad_norm": 0.19503540018100968,
|
|
"learning_rate": 4.499521541879508e-06,
|
|
"loss": 0.7316,
|
|
"step": 503
|
|
},
|
|
{
|
|
"epoch": 0.546748932130992,
|
|
"grad_norm": 0.22534483356707935,
|
|
"learning_rate": 4.48202291137026e-06,
|
|
"loss": 0.7305,
|
|
"step": 504
|
|
},
|
|
{
|
|
"epoch": 0.5478337514407756,
|
|
"grad_norm": 0.21069421706833882,
|
|
"learning_rate": 4.4645306915828025e-06,
|
|
"loss": 0.7303,
|
|
"step": 505
|
|
},
|
|
{
|
|
"epoch": 0.5489185707505594,
|
|
"grad_norm": 0.2112810124760799,
|
|
"learning_rate": 4.447045099008863e-06,
|
|
"loss": 0.7275,
|
|
"step": 506
|
|
},
|
|
{
|
|
"epoch": 0.550003390060343,
|
|
"grad_norm": 0.19763267259678646,
|
|
"learning_rate": 4.429566350058146e-06,
|
|
"loss": 0.7303,
|
|
"step": 507
|
|
},
|
|
{
|
|
"epoch": 0.5510882093701268,
|
|
"grad_norm": 0.20914686642317265,
|
|
"learning_rate": 4.412094661055658e-06,
|
|
"loss": 0.7335,
|
|
"step": 508
|
|
},
|
|
{
|
|
"epoch": 0.5521730286799105,
|
|
"grad_norm": 0.19650544428652253,
|
|
"learning_rate": 4.394630248239029e-06,
|
|
"loss": 0.7328,
|
|
"step": 509
|
|
},
|
|
{
|
|
"epoch": 0.5532578479896942,
|
|
"grad_norm": 0.22274889110339527,
|
|
"learning_rate": 4.377173327755832e-06,
|
|
"loss": 0.7293,
|
|
"step": 510
|
|
},
|
|
{
|
|
"epoch": 0.5532578479896942,
|
|
"eval_loss": 0.7311838865280151,
|
|
"eval_runtime": 109.0794,
|
|
"eval_samples_per_second": 87.404,
|
|
"eval_steps_per_second": 0.688,
|
|
"step": 510
|
|
},
|
|
{
|
|
"epoch": 0.5543426672994779,
|
|
"grad_norm": 0.19383901815984803,
|
|
"learning_rate": 4.359724115660915e-06,
|
|
"loss": 0.7331,
|
|
"step": 511
|
|
},
|
|
{
|
|
"epoch": 0.5554274866092617,
|
|
"grad_norm": 0.20371405109054755,
|
|
"learning_rate": 4.3422828279137245e-06,
|
|
"loss": 0.7291,
|
|
"step": 512
|
|
},
|
|
{
|
|
"epoch": 0.5565123059190453,
|
|
"grad_norm": 0.21471981276385824,
|
|
"learning_rate": 4.324849680375625e-06,
|
|
"loss": 0.729,
|
|
"step": 513
|
|
},
|
|
{
|
|
"epoch": 0.5575971252288291,
|
|
"grad_norm": 0.18335095375136745,
|
|
"learning_rate": 4.307424888807242e-06,
|
|
"loss": 0.7346,
|
|
"step": 514
|
|
},
|
|
{
|
|
"epoch": 0.5586819445386128,
|
|
"grad_norm": 0.18903106969399844,
|
|
"learning_rate": 4.290008668865778e-06,
|
|
"loss": 0.7311,
|
|
"step": 515
|
|
},
|
|
{
|
|
"epoch": 0.5597667638483965,
|
|
"grad_norm": 0.20951505646038296,
|
|
"learning_rate": 4.272601236102353e-06,
|
|
"loss": 0.7308,
|
|
"step": 516
|
|
},
|
|
{
|
|
"epoch": 0.5608515831581802,
|
|
"grad_norm": 0.19210850860939485,
|
|
"learning_rate": 4.2552028059593294e-06,
|
|
"loss": 0.7299,
|
|
"step": 517
|
|
},
|
|
{
|
|
"epoch": 0.561936402467964,
|
|
"grad_norm": 0.21978159580742676,
|
|
"learning_rate": 4.2378135937676515e-06,
|
|
"loss": 0.7291,
|
|
"step": 518
|
|
},
|
|
{
|
|
"epoch": 0.5630212217777476,
|
|
"grad_norm": 0.1750060487501776,
|
|
"learning_rate": 4.220433814744179e-06,
|
|
"loss": 0.731,
|
|
"step": 519
|
|
},
|
|
{
|
|
"epoch": 0.5641060410875314,
|
|
"grad_norm": 0.21984370214536691,
|
|
"learning_rate": 4.203063683989017e-06,
|
|
"loss": 0.7302,
|
|
"step": 520
|
|
},
|
|
{
|
|
"epoch": 0.5641060410875314,
|
|
"eval_loss": 0.7307312488555908,
|
|
"eval_runtime": 108.9757,
|
|
"eval_samples_per_second": 87.487,
|
|
"eval_steps_per_second": 0.688,
|
|
"step": 520
|
|
},
|
|
{
|
|
"epoch": 0.565190860397315,
|
|
"grad_norm": 0.18038329794134672,
|
|
"learning_rate": 4.185703416482867e-06,
|
|
"loss": 0.7278,
|
|
"step": 521
|
|
},
|
|
{
|
|
"epoch": 0.5662756797070988,
|
|
"grad_norm": 0.20839672674225715,
|
|
"learning_rate": 4.1683532270843505e-06,
|
|
"loss": 0.7291,
|
|
"step": 522
|
|
},
|
|
{
|
|
"epoch": 0.5673604990168825,
|
|
"grad_norm": 0.1933030055299005,
|
|
"learning_rate": 4.151013330527364e-06,
|
|
"loss": 0.7318,
|
|
"step": 523
|
|
},
|
|
{
|
|
"epoch": 0.5684453183266662,
|
|
"grad_norm": 0.20147615340390618,
|
|
"learning_rate": 4.133683941418411e-06,
|
|
"loss": 0.731,
|
|
"step": 524
|
|
},
|
|
{
|
|
"epoch": 0.5695301376364499,
|
|
"grad_norm": 0.1733377774068175,
|
|
"learning_rate": 4.116365274233952e-06,
|
|
"loss": 0.7316,
|
|
"step": 525
|
|
},
|
|
{
|
|
"epoch": 0.5706149569462337,
|
|
"grad_norm": 0.19466724290497228,
|
|
"learning_rate": 4.099057543317749e-06,
|
|
"loss": 0.7294,
|
|
"step": 526
|
|
},
|
|
{
|
|
"epoch": 0.5716997762560173,
|
|
"grad_norm": 0.17658152130537005,
|
|
"learning_rate": 4.081760962878209e-06,
|
|
"loss": 0.7325,
|
|
"step": 527
|
|
},
|
|
{
|
|
"epoch": 0.5727845955658011,
|
|
"grad_norm": 0.17454394154659186,
|
|
"learning_rate": 4.064475746985738e-06,
|
|
"loss": 0.7339,
|
|
"step": 528
|
|
},
|
|
{
|
|
"epoch": 0.5738694148755847,
|
|
"grad_norm": 0.16455282945936323,
|
|
"learning_rate": 4.047202109570086e-06,
|
|
"loss": 0.726,
|
|
"step": 529
|
|
},
|
|
{
|
|
"epoch": 0.5749542341853685,
|
|
"grad_norm": 0.1903687662585094,
|
|
"learning_rate": 4.029940264417708e-06,
|
|
"loss": 0.7293,
|
|
"step": 530
|
|
},
|
|
{
|
|
"epoch": 0.5749542341853685,
|
|
"eval_loss": 0.730327308177948,
|
|
"eval_runtime": 108.7704,
|
|
"eval_samples_per_second": 87.653,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 530
|
|
},
|
|
{
|
|
"epoch": 0.5760390534951522,
|
|
"grad_norm": 0.1961049848257356,
|
|
"learning_rate": 4.012690425169104e-06,
|
|
"loss": 0.7317,
|
|
"step": 531
|
|
},
|
|
{
|
|
"epoch": 0.5771238728049359,
|
|
"grad_norm": 0.1866576105200039,
|
|
"learning_rate": 3.995452805316195e-06,
|
|
"loss": 0.7274,
|
|
"step": 532
|
|
},
|
|
{
|
|
"epoch": 0.5782086921147196,
|
|
"grad_norm": 0.2133702344554463,
|
|
"learning_rate": 3.978227618199657e-06,
|
|
"loss": 0.73,
|
|
"step": 533
|
|
},
|
|
{
|
|
"epoch": 0.5792935114245034,
|
|
"grad_norm": 0.2150953120623853,
|
|
"learning_rate": 3.961015077006301e-06,
|
|
"loss": 0.7328,
|
|
"step": 534
|
|
},
|
|
{
|
|
"epoch": 0.580378330734287,
|
|
"grad_norm": 0.1863251132908174,
|
|
"learning_rate": 3.943815394766426e-06,
|
|
"loss": 0.7286,
|
|
"step": 535
|
|
},
|
|
{
|
|
"epoch": 0.5814631500440708,
|
|
"grad_norm": 0.199019936612921,
|
|
"learning_rate": 3.926628784351175e-06,
|
|
"loss": 0.7296,
|
|
"step": 536
|
|
},
|
|
{
|
|
"epoch": 0.5825479693538544,
|
|
"grad_norm": 0.20157094830069663,
|
|
"learning_rate": 3.90945545846992e-06,
|
|
"loss": 0.7289,
|
|
"step": 537
|
|
},
|
|
{
|
|
"epoch": 0.5836327886636382,
|
|
"grad_norm": 0.2007918930762589,
|
|
"learning_rate": 3.892295629667604e-06,
|
|
"loss": 0.728,
|
|
"step": 538
|
|
},
|
|
{
|
|
"epoch": 0.584717607973422,
|
|
"grad_norm": 0.2249399096096621,
|
|
"learning_rate": 3.875149510322137e-06,
|
|
"loss": 0.7308,
|
|
"step": 539
|
|
},
|
|
{
|
|
"epoch": 0.5858024272832056,
|
|
"grad_norm": 0.1896024018851033,
|
|
"learning_rate": 3.8580173126417455e-06,
|
|
"loss": 0.7298,
|
|
"step": 540
|
|
},
|
|
{
|
|
"epoch": 0.5858024272832056,
|
|
"eval_loss": 0.7299535274505615,
|
|
"eval_runtime": 109.1123,
|
|
"eval_samples_per_second": 87.378,
|
|
"eval_steps_per_second": 0.687,
|
|
"step": 540
|
|
},
|
|
{
|
|
"epoch": 0.5868872465929894,
|
|
"grad_norm": 0.22458413196239288,
|
|
"learning_rate": 3.840899248662358e-06,
|
|
"loss": 0.7293,
|
|
"step": 541
|
|
},
|
|
{
|
|
"epoch": 0.5879720659027731,
|
|
"grad_norm": 0.18155874758546836,
|
|
"learning_rate": 3.823795530244982e-06,
|
|
"loss": 0.7323,
|
|
"step": 542
|
|
},
|
|
{
|
|
"epoch": 0.5890568852125568,
|
|
"grad_norm": 0.2093205260031526,
|
|
"learning_rate": 3.8067063690730672e-06,
|
|
"loss": 0.7307,
|
|
"step": 543
|
|
},
|
|
{
|
|
"epoch": 0.5901417045223405,
|
|
"grad_norm": 0.18577649194505946,
|
|
"learning_rate": 3.7896319766499073e-06,
|
|
"loss": 0.7269,
|
|
"step": 544
|
|
},
|
|
{
|
|
"epoch": 0.5912265238321243,
|
|
"grad_norm": 0.20060193258833295,
|
|
"learning_rate": 3.7725725642960047e-06,
|
|
"loss": 0.7325,
|
|
"step": 545
|
|
},
|
|
{
|
|
"epoch": 0.5923113431419079,
|
|
"grad_norm": 0.19343315743346895,
|
|
"learning_rate": 3.755528343146465e-06,
|
|
"loss": 0.7291,
|
|
"step": 546
|
|
},
|
|
{
|
|
"epoch": 0.5933961624516917,
|
|
"grad_norm": 0.19577245906897306,
|
|
"learning_rate": 3.7384995241483767e-06,
|
|
"loss": 0.7326,
|
|
"step": 547
|
|
},
|
|
{
|
|
"epoch": 0.5944809817614753,
|
|
"grad_norm": 0.2044579113170833,
|
|
"learning_rate": 3.7214863180582085e-06,
|
|
"loss": 0.7307,
|
|
"step": 548
|
|
},
|
|
{
|
|
"epoch": 0.5955658010712591,
|
|
"grad_norm": 0.19029031334330762,
|
|
"learning_rate": 3.7044889354391934e-06,
|
|
"loss": 0.7282,
|
|
"step": 549
|
|
},
|
|
{
|
|
"epoch": 0.5966506203810428,
|
|
"grad_norm": 0.20035245146316058,
|
|
"learning_rate": 3.687507586658726e-06,
|
|
"loss": 0.7275,
|
|
"step": 550
|
|
},
|
|
{
|
|
"epoch": 0.5966506203810428,
|
|
"eval_loss": 0.7295899391174316,
|
|
"eval_runtime": 108.8135,
|
|
"eval_samples_per_second": 87.618,
|
|
"eval_steps_per_second": 0.689,
|
|
"step": 550
|
|
},
|
|
{
|
|
"epoch": 0.5977354396908265,
|
|
"grad_norm": 0.20508091155864092,
|
|
"learning_rate": 3.6705424818857636e-06,
|
|
"loss": 0.7292,
|
|
"step": 551
|
|
},
|
|
{
|
|
"epoch": 0.5988202590006102,
|
|
"grad_norm": 0.17833197540734563,
|
|
"learning_rate": 3.6535938310882124e-06,
|
|
"loss": 0.7292,
|
|
"step": 552
|
|
},
|
|
{
|
|
"epoch": 0.599905078310394,
|
|
"grad_norm": 0.22585346874491902,
|
|
"learning_rate": 3.6366618440303436e-06,
|
|
"loss": 0.7307,
|
|
"step": 553
|
|
},
|
|
{
|
|
"epoch": 0.6009898976201776,
|
|
"grad_norm": 0.18255823625646447,
|
|
"learning_rate": 3.619746730270185e-06,
|
|
"loss": 0.73,
|
|
"step": 554
|
|
},
|
|
{
|
|
"epoch": 0.6020747169299614,
|
|
"grad_norm": 0.18720416109346225,
|
|
"learning_rate": 3.6028486991569376e-06,
|
|
"loss": 0.7326,
|
|
"step": 555
|
|
},
|
|
{
|
|
"epoch": 0.603159536239745,
|
|
"grad_norm": 0.19669162634777657,
|
|
"learning_rate": 3.585967959828375e-06,
|
|
"loss": 0.7322,
|
|
"step": 556
|
|
},
|
|
{
|
|
"epoch": 0.6042443555495288,
|
|
"grad_norm": 0.18364689905375967,
|
|
"learning_rate": 3.569104721208262e-06,
|
|
"loss": 0.7333,
|
|
"step": 557
|
|
},
|
|
{
|
|
"epoch": 0.6053291748593125,
|
|
"grad_norm": 0.19067909572323474,
|
|
"learning_rate": 3.5522591920037698e-06,
|
|
"loss": 0.7316,
|
|
"step": 558
|
|
},
|
|
{
|
|
"epoch": 0.6064139941690962,
|
|
"grad_norm": 0.1810431495158628,
|
|
"learning_rate": 3.5354315807028826e-06,
|
|
"loss": 0.7311,
|
|
"step": 559
|
|
},
|
|
{
|
|
"epoch": 0.6074988134788799,
|
|
"grad_norm": 0.18678300295802247,
|
|
"learning_rate": 3.518622095571831e-06,
|
|
"loss": 0.7295,
|
|
"step": 560
|
|
},
|
|
{
|
|
"epoch": 0.6074988134788799,
|
|
"eval_loss": 0.7291638255119324,
|
|
"eval_runtime": 108.7906,
|
|
"eval_samples_per_second": 87.636,
|
|
"eval_steps_per_second": 0.689,
|
|
"step": 560
|
|
},
|
|
{
|
|
"epoch": 0.6085836327886637,
|
|
"grad_norm": 0.17489948387314008,
|
|
"learning_rate": 3.501830944652504e-06,
|
|
"loss": 0.7306,
|
|
"step": 561
|
|
},
|
|
{
|
|
"epoch": 0.6096684520984473,
|
|
"grad_norm": 0.19211293958904063,
|
|
"learning_rate": 3.4850583357598805e-06,
|
|
"loss": 0.7286,
|
|
"step": 562
|
|
},
|
|
{
|
|
"epoch": 0.6107532714082311,
|
|
"grad_norm": 0.19873116945009495,
|
|
"learning_rate": 3.4683044764794516e-06,
|
|
"loss": 0.7275,
|
|
"step": 563
|
|
},
|
|
{
|
|
"epoch": 0.6118380907180148,
|
|
"grad_norm": 0.18876393571192043,
|
|
"learning_rate": 3.451569574164658e-06,
|
|
"loss": 0.7316,
|
|
"step": 564
|
|
},
|
|
{
|
|
"epoch": 0.6129229100277985,
|
|
"grad_norm": 0.18545841727248993,
|
|
"learning_rate": 3.4348538359343187e-06,
|
|
"loss": 0.7276,
|
|
"step": 565
|
|
},
|
|
{
|
|
"epoch": 0.6140077293375822,
|
|
"grad_norm": 0.20096108577103247,
|
|
"learning_rate": 3.4181574686700687e-06,
|
|
"loss": 0.7264,
|
|
"step": 566
|
|
},
|
|
{
|
|
"epoch": 0.615092548647366,
|
|
"grad_norm": 0.18696572286488702,
|
|
"learning_rate": 3.401480679013801e-06,
|
|
"loss": 0.7317,
|
|
"step": 567
|
|
},
|
|
{
|
|
"epoch": 0.6161773679571496,
|
|
"grad_norm": 0.19177392876558408,
|
|
"learning_rate": 3.3848236733651034e-06,
|
|
"loss": 0.7309,
|
|
"step": 568
|
|
},
|
|
{
|
|
"epoch": 0.6172621872669334,
|
|
"grad_norm": 0.17818918419718113,
|
|
"learning_rate": 3.3681866578787124e-06,
|
|
"loss": 0.7303,
|
|
"step": 569
|
|
},
|
|
{
|
|
"epoch": 0.618347006576717,
|
|
"grad_norm": 0.20003253823670258,
|
|
"learning_rate": 3.3515698384619543e-06,
|
|
"loss": 0.7282,
|
|
"step": 570
|
|
},
|
|
{
|
|
"epoch": 0.618347006576717,
|
|
"eval_loss": 0.7288394570350647,
|
|
"eval_runtime": 108.7655,
|
|
"eval_samples_per_second": 87.656,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 570
|
|
},
|
|
{
|
|
"epoch": 0.6194318258865008,
|
|
"grad_norm": 0.18488999710186219,
|
|
"learning_rate": 3.3349734207722e-06,
|
|
"loss": 0.7304,
|
|
"step": 571
|
|
},
|
|
{
|
|
"epoch": 0.6205166451962845,
|
|
"grad_norm": 0.17012657974558315,
|
|
"learning_rate": 3.318397610214319e-06,
|
|
"loss": 0.7307,
|
|
"step": 572
|
|
},
|
|
{
|
|
"epoch": 0.6216014645060682,
|
|
"grad_norm": 0.19004829438050294,
|
|
"learning_rate": 3.3018426119381364e-06,
|
|
"loss": 0.7284,
|
|
"step": 573
|
|
},
|
|
{
|
|
"epoch": 0.6226862838158519,
|
|
"grad_norm": 0.1691948235105627,
|
|
"learning_rate": 3.285308630835903e-06,
|
|
"loss": 0.7318,
|
|
"step": 574
|
|
},
|
|
{
|
|
"epoch": 0.6237711031256357,
|
|
"grad_norm": 0.18402142217713774,
|
|
"learning_rate": 3.2687958715397373e-06,
|
|
"loss": 0.7279,
|
|
"step": 575
|
|
},
|
|
{
|
|
"epoch": 0.6248559224354193,
|
|
"grad_norm": 0.17806533270148225,
|
|
"learning_rate": 3.2523045384191186e-06,
|
|
"loss": 0.7296,
|
|
"step": 576
|
|
},
|
|
{
|
|
"epoch": 0.6259407417452031,
|
|
"grad_norm": 0.1706597248487234,
|
|
"learning_rate": 3.235834835578341e-06,
|
|
"loss": 0.7255,
|
|
"step": 577
|
|
},
|
|
{
|
|
"epoch": 0.6270255610549867,
|
|
"grad_norm": 0.2051010802159253,
|
|
"learning_rate": 3.2193869668539947e-06,
|
|
"loss": 0.7269,
|
|
"step": 578
|
|
},
|
|
{
|
|
"epoch": 0.6281103803647705,
|
|
"grad_norm": 0.1709919049540414,
|
|
"learning_rate": 3.202961135812437e-06,
|
|
"loss": 0.727,
|
|
"step": 579
|
|
},
|
|
{
|
|
"epoch": 0.6291951996745542,
|
|
"grad_norm": 0.1967728234250428,
|
|
"learning_rate": 3.1865575457472797e-06,
|
|
"loss": 0.7307,
|
|
"step": 580
|
|
},
|
|
{
|
|
"epoch": 0.6291951996745542,
|
|
"eval_loss": 0.728428065776825,
|
|
"eval_runtime": 108.8598,
|
|
"eval_samples_per_second": 87.581,
|
|
"eval_steps_per_second": 0.689,
|
|
"step": 580
|
|
},
|
|
{
|
|
"epoch": 0.6302800189843379,
|
|
"grad_norm": 0.17127342321162956,
|
|
"learning_rate": 3.1701763996768744e-06,
|
|
"loss": 0.727,
|
|
"step": 581
|
|
},
|
|
{
|
|
"epoch": 0.6313648382941216,
|
|
"grad_norm": 0.1807860097189854,
|
|
"learning_rate": 3.1538179003417836e-06,
|
|
"loss": 0.7321,
|
|
"step": 582
|
|
},
|
|
{
|
|
"epoch": 0.6324496576039054,
|
|
"grad_norm": 0.18745528570763353,
|
|
"learning_rate": 3.137482250202298e-06,
|
|
"loss": 0.7282,
|
|
"step": 583
|
|
},
|
|
{
|
|
"epoch": 0.633534476913689,
|
|
"grad_norm": 0.17447454296730197,
|
|
"learning_rate": 3.121169651435903e-06,
|
|
"loss": 0.7265,
|
|
"step": 584
|
|
},
|
|
{
|
|
"epoch": 0.6346192962234728,
|
|
"grad_norm": 0.1972301331018418,
|
|
"learning_rate": 3.104880305934799e-06,
|
|
"loss": 0.7321,
|
|
"step": 585
|
|
},
|
|
{
|
|
"epoch": 0.6357041155332565,
|
|
"grad_norm": 0.17005762200419927,
|
|
"learning_rate": 3.088614415303387e-06,
|
|
"loss": 0.7249,
|
|
"step": 586
|
|
},
|
|
{
|
|
"epoch": 0.6367889348430402,
|
|
"grad_norm": 0.19135582669856777,
|
|
"learning_rate": 3.0723721808557857e-06,
|
|
"loss": 0.7268,
|
|
"step": 587
|
|
},
|
|
{
|
|
"epoch": 0.6378737541528239,
|
|
"grad_norm": 0.18493750751814608,
|
|
"learning_rate": 3.0561538036133275e-06,
|
|
"loss": 0.7281,
|
|
"step": 588
|
|
},
|
|
{
|
|
"epoch": 0.6389585734626076,
|
|
"grad_norm": 0.17447235043525056,
|
|
"learning_rate": 3.039959484302083e-06,
|
|
"loss": 0.7271,
|
|
"step": 589
|
|
},
|
|
{
|
|
"epoch": 0.6400433927723913,
|
|
"grad_norm": 0.1693781527426387,
|
|
"learning_rate": 3.0237894233503697e-06,
|
|
"loss": 0.7289,
|
|
"step": 590
|
|
},
|
|
{
|
|
"epoch": 0.6400433927723913,
|
|
"eval_loss": 0.7281356453895569,
|
|
"eval_runtime": 108.6797,
|
|
"eval_samples_per_second": 87.726,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 590
|
|
},
|
|
{
|
|
"epoch": 0.6411282120821751,
|
|
"grad_norm": 0.18690524500684685,
|
|
"learning_rate": 3.007643820886267e-06,
|
|
"loss": 0.7265,
|
|
"step": 591
|
|
},
|
|
{
|
|
"epoch": 0.6422130313919587,
|
|
"grad_norm": 0.17414038846144117,
|
|
"learning_rate": 2.991522876735154e-06,
|
|
"loss": 0.7312,
|
|
"step": 592
|
|
},
|
|
{
|
|
"epoch": 0.6432978507017425,
|
|
"grad_norm": 0.1843906904490678,
|
|
"learning_rate": 2.975426790417218e-06,
|
|
"loss": 0.726,
|
|
"step": 593
|
|
},
|
|
{
|
|
"epoch": 0.6443826700115262,
|
|
"grad_norm": 0.17489370264444826,
|
|
"learning_rate": 2.959355761145001e-06,
|
|
"loss": 0.7245,
|
|
"step": 594
|
|
},
|
|
{
|
|
"epoch": 0.6454674893213099,
|
|
"grad_norm": 0.20227377539437683,
|
|
"learning_rate": 2.9433099878209238e-06,
|
|
"loss": 0.7316,
|
|
"step": 595
|
|
},
|
|
{
|
|
"epoch": 0.6465523086310936,
|
|
"grad_norm": 0.1539367971852112,
|
|
"learning_rate": 2.9272896690348283e-06,
|
|
"loss": 0.7273,
|
|
"step": 596
|
|
},
|
|
{
|
|
"epoch": 0.6476371279408774,
|
|
"grad_norm": 0.1963469543809783,
|
|
"learning_rate": 2.911295003061526e-06,
|
|
"loss": 0.7259,
|
|
"step": 597
|
|
},
|
|
{
|
|
"epoch": 0.648721947250661,
|
|
"grad_norm": 0.1698731707765151,
|
|
"learning_rate": 2.8953261878583263e-06,
|
|
"loss": 0.7278,
|
|
"step": 598
|
|
},
|
|
{
|
|
"epoch": 0.6498067665604448,
|
|
"grad_norm": 0.17972192448266683,
|
|
"learning_rate": 2.8793834210626036e-06,
|
|
"loss": 0.73,
|
|
"step": 599
|
|
},
|
|
{
|
|
"epoch": 0.6508915858702284,
|
|
"grad_norm": 0.18318276082273133,
|
|
"learning_rate": 2.8634668999893477e-06,
|
|
"loss": 0.7277,
|
|
"step": 600
|
|
},
|
|
{
|
|
"epoch": 0.6508915858702284,
|
|
"eval_loss": 0.7278279066085815,
|
|
"eval_runtime": 108.7109,
|
|
"eval_samples_per_second": 87.7,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 600
|
|
},
|
|
{
|
|
"epoch": 0.6519764051800122,
|
|
"grad_norm": 0.1673113201990224,
|
|
"learning_rate": 2.847576821628716e-06,
|
|
"loss": 0.7248,
|
|
"step": 601
|
|
},
|
|
{
|
|
"epoch": 0.6530612244897959,
|
|
"grad_norm": 0.16941616256775213,
|
|
"learning_rate": 2.8317133826435968e-06,
|
|
"loss": 0.7267,
|
|
"step": 602
|
|
},
|
|
{
|
|
"epoch": 0.6541460437995796,
|
|
"grad_norm": 0.16441880631489086,
|
|
"learning_rate": 2.815876779367181e-06,
|
|
"loss": 0.7284,
|
|
"step": 603
|
|
},
|
|
{
|
|
"epoch": 0.6552308631093633,
|
|
"grad_norm": 0.16675452754586417,
|
|
"learning_rate": 2.8000672078005277e-06,
|
|
"loss": 0.7263,
|
|
"step": 604
|
|
},
|
|
{
|
|
"epoch": 0.6563156824191471,
|
|
"grad_norm": 0.163123864912065,
|
|
"learning_rate": 2.784284863610138e-06,
|
|
"loss": 0.7292,
|
|
"step": 605
|
|
},
|
|
{
|
|
"epoch": 0.6574005017289307,
|
|
"grad_norm": 0.16767051474583367,
|
|
"learning_rate": 2.7685299421255373e-06,
|
|
"loss": 0.7236,
|
|
"step": 606
|
|
},
|
|
{
|
|
"epoch": 0.6584853210387145,
|
|
"grad_norm": 0.1626462064287084,
|
|
"learning_rate": 2.75280263833685e-06,
|
|
"loss": 0.729,
|
|
"step": 607
|
|
},
|
|
{
|
|
"epoch": 0.6595701403484983,
|
|
"grad_norm": 0.170111812715493,
|
|
"learning_rate": 2.737103146892395e-06,
|
|
"loss": 0.7257,
|
|
"step": 608
|
|
},
|
|
{
|
|
"epoch": 0.6606549596582819,
|
|
"grad_norm": 0.15597077857965097,
|
|
"learning_rate": 2.7214316620962727e-06,
|
|
"loss": 0.7267,
|
|
"step": 609
|
|
},
|
|
{
|
|
"epoch": 0.6617397789680657,
|
|
"grad_norm": 0.16255812984452953,
|
|
"learning_rate": 2.705788377905961e-06,
|
|
"loss": 0.7246,
|
|
"step": 610
|
|
},
|
|
{
|
|
"epoch": 0.6617397789680657,
|
|
"eval_loss": 0.727541446685791,
|
|
"eval_runtime": 108.7603,
|
|
"eval_samples_per_second": 87.661,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 610
|
|
},
|
|
{
|
|
"epoch": 0.6628245982778493,
|
|
"grad_norm": 0.16461566220431453,
|
|
"learning_rate": 2.69017348792991e-06,
|
|
"loss": 0.7278,
|
|
"step": 611
|
|
},
|
|
{
|
|
"epoch": 0.6639094175876331,
|
|
"grad_norm": 0.15791454720333606,
|
|
"learning_rate": 2.674587185425155e-06,
|
|
"loss": 0.7275,
|
|
"step": 612
|
|
},
|
|
{
|
|
"epoch": 0.6649942368974168,
|
|
"grad_norm": 0.17312188406262885,
|
|
"learning_rate": 2.6590296632949157e-06,
|
|
"loss": 0.7305,
|
|
"step": 613
|
|
},
|
|
{
|
|
"epoch": 0.6660790562072005,
|
|
"grad_norm": 0.16603472085953,
|
|
"learning_rate": 2.6435011140862167e-06,
|
|
"loss": 0.7275,
|
|
"step": 614
|
|
},
|
|
{
|
|
"epoch": 0.6671638755169842,
|
|
"grad_norm": 0.17122376858054306,
|
|
"learning_rate": 2.6280017299874984e-06,
|
|
"loss": 0.7267,
|
|
"step": 615
|
|
},
|
|
{
|
|
"epoch": 0.668248694826768,
|
|
"grad_norm": 0.16175628153447158,
|
|
"learning_rate": 2.6125317028262383e-06,
|
|
"loss": 0.7273,
|
|
"step": 616
|
|
},
|
|
{
|
|
"epoch": 0.6693335141365516,
|
|
"grad_norm": 0.17429091470871394,
|
|
"learning_rate": 2.5970912240665815e-06,
|
|
"loss": 0.7288,
|
|
"step": 617
|
|
},
|
|
{
|
|
"epoch": 0.6704183334463354,
|
|
"grad_norm": 0.1679703554640434,
|
|
"learning_rate": 2.5816804848069693e-06,
|
|
"loss": 0.7291,
|
|
"step": 618
|
|
},
|
|
{
|
|
"epoch": 0.671503152756119,
|
|
"grad_norm": 0.1630148155956674,
|
|
"learning_rate": 2.5662996757777716e-06,
|
|
"loss": 0.7287,
|
|
"step": 619
|
|
},
|
|
{
|
|
"epoch": 0.6725879720659028,
|
|
"grad_norm": 0.16588068506650458,
|
|
"learning_rate": 2.550948987338929e-06,
|
|
"loss": 0.729,
|
|
"step": 620
|
|
},
|
|
{
|
|
"epoch": 0.6725879720659028,
|
|
"eval_loss": 0.7272528409957886,
|
|
"eval_runtime": 108.8752,
|
|
"eval_samples_per_second": 87.568,
|
|
"eval_steps_per_second": 0.689,
|
|
"step": 620
|
|
},
|
|
{
|
|
"epoch": 0.6736727913756865,
|
|
"grad_norm": 0.16791141353676373,
|
|
"learning_rate": 2.5356286094775943e-06,
|
|
"loss": 0.7295,
|
|
"step": 621
|
|
},
|
|
{
|
|
"epoch": 0.6747576106854702,
|
|
"grad_norm": 0.16934465370604204,
|
|
"learning_rate": 2.520338731805785e-06,
|
|
"loss": 0.7283,
|
|
"step": 622
|
|
},
|
|
{
|
|
"epoch": 0.6758424299952539,
|
|
"grad_norm": 0.1643142718072723,
|
|
"learning_rate": 2.505079543558031e-06,
|
|
"loss": 0.7295,
|
|
"step": 623
|
|
},
|
|
{
|
|
"epoch": 0.6769272493050377,
|
|
"grad_norm": 0.16261768382002517,
|
|
"learning_rate": 2.4898512335890425e-06,
|
|
"loss": 0.7264,
|
|
"step": 624
|
|
},
|
|
{
|
|
"epoch": 0.6780120686148213,
|
|
"grad_norm": 0.17147132095368184,
|
|
"learning_rate": 2.474653990371356e-06,
|
|
"loss": 0.7287,
|
|
"step": 625
|
|
},
|
|
{
|
|
"epoch": 0.6790968879246051,
|
|
"grad_norm": 0.1611487106075524,
|
|
"learning_rate": 2.4594880019930194e-06,
|
|
"loss": 0.7285,
|
|
"step": 626
|
|
},
|
|
{
|
|
"epoch": 0.6801817072343888,
|
|
"grad_norm": 0.17446583917527791,
|
|
"learning_rate": 2.4443534561552543e-06,
|
|
"loss": 0.7226,
|
|
"step": 627
|
|
},
|
|
{
|
|
"epoch": 0.6812665265441725,
|
|
"grad_norm": 0.16629465774257843,
|
|
"learning_rate": 2.429250540170135e-06,
|
|
"loss": 0.7292,
|
|
"step": 628
|
|
},
|
|
{
|
|
"epoch": 0.6823513458539562,
|
|
"grad_norm": 0.17665908778521658,
|
|
"learning_rate": 2.4141794409582713e-06,
|
|
"loss": 0.7276,
|
|
"step": 629
|
|
},
|
|
{
|
|
"epoch": 0.68343616516374,
|
|
"grad_norm": 0.16632825619484914,
|
|
"learning_rate": 2.3991403450464896e-06,
|
|
"loss": 0.7244,
|
|
"step": 630
|
|
},
|
|
{
|
|
"epoch": 0.68343616516374,
|
|
"eval_loss": 0.7270001173019409,
|
|
"eval_runtime": 108.8695,
|
|
"eval_samples_per_second": 87.573,
|
|
"eval_steps_per_second": 0.689,
|
|
"step": 630
|
|
},
|
|
{
|
|
"epoch": 0.6845209844735236,
|
|
"grad_norm": 0.18224111565421297,
|
|
"learning_rate": 2.384133438565533e-06,
|
|
"loss": 0.7263,
|
|
"step": 631
|
|
},
|
|
{
|
|
"epoch": 0.6856058037833074,
|
|
"grad_norm": 0.17069033066568248,
|
|
"learning_rate": 2.3691589072477527e-06,
|
|
"loss": 0.7272,
|
|
"step": 632
|
|
},
|
|
{
|
|
"epoch": 0.686690623093091,
|
|
"grad_norm": 0.1636940508571279,
|
|
"learning_rate": 2.354216936424812e-06,
|
|
"loss": 0.7269,
|
|
"step": 633
|
|
},
|
|
{
|
|
"epoch": 0.6877754424028748,
|
|
"grad_norm": 0.15914737968622805,
|
|
"learning_rate": 2.3393077110253838e-06,
|
|
"loss": 0.7274,
|
|
"step": 634
|
|
},
|
|
{
|
|
"epoch": 0.6888602617126585,
|
|
"grad_norm": 0.1722425292196802,
|
|
"learning_rate": 2.3244314155728758e-06,
|
|
"loss": 0.7273,
|
|
"step": 635
|
|
},
|
|
{
|
|
"epoch": 0.6899450810224422,
|
|
"grad_norm": 0.16907875976639417,
|
|
"learning_rate": 2.309588234183137e-06,
|
|
"loss": 0.7254,
|
|
"step": 636
|
|
},
|
|
{
|
|
"epoch": 0.6910299003322259,
|
|
"grad_norm": 0.1633749318860212,
|
|
"learning_rate": 2.2947783505621813e-06,
|
|
"loss": 0.7295,
|
|
"step": 637
|
|
},
|
|
{
|
|
"epoch": 0.6921147196420097,
|
|
"grad_norm": 0.1654906038637897,
|
|
"learning_rate": 2.280001948003916e-06,
|
|
"loss": 0.7256,
|
|
"step": 638
|
|
},
|
|
{
|
|
"epoch": 0.6931995389517933,
|
|
"grad_norm": 0.16610707761149088,
|
|
"learning_rate": 2.265259209387867e-06,
|
|
"loss": 0.7226,
|
|
"step": 639
|
|
},
|
|
{
|
|
"epoch": 0.6942843582615771,
|
|
"grad_norm": 0.15508587899367243,
|
|
"learning_rate": 2.2505503171769233e-06,
|
|
"loss": 0.7272,
|
|
"step": 640
|
|
},
|
|
{
|
|
"epoch": 0.6942843582615771,
|
|
"eval_loss": 0.7267536520957947,
|
|
"eval_runtime": 108.7887,
|
|
"eval_samples_per_second": 87.638,
|
|
"eval_steps_per_second": 0.689,
|
|
"step": 640
|
|
},
|
|
{
|
|
"epoch": 0.6953691775713607,
|
|
"grad_norm": 0.16476646065490425,
|
|
"learning_rate": 2.2358754534150752e-06,
|
|
"loss": 0.726,
|
|
"step": 641
|
|
},
|
|
{
|
|
"epoch": 0.6964539968811445,
|
|
"grad_norm": 0.1694740328326384,
|
|
"learning_rate": 2.221234799725161e-06,
|
|
"loss": 0.7264,
|
|
"step": 642
|
|
},
|
|
{
|
|
"epoch": 0.6975388161909282,
|
|
"grad_norm": 0.15804754433564264,
|
|
"learning_rate": 2.206628537306621e-06,
|
|
"loss": 0.7288,
|
|
"step": 643
|
|
},
|
|
{
|
|
"epoch": 0.6986236355007119,
|
|
"grad_norm": 0.1630387039532913,
|
|
"learning_rate": 2.1920568469332458e-06,
|
|
"loss": 0.7252,
|
|
"step": 644
|
|
},
|
|
{
|
|
"epoch": 0.6997084548104956,
|
|
"grad_norm": 0.1615046300021262,
|
|
"learning_rate": 2.17751990895096e-06,
|
|
"loss": 0.7268,
|
|
"step": 645
|
|
},
|
|
{
|
|
"epoch": 0.7007932741202794,
|
|
"grad_norm": 0.16947378751426528,
|
|
"learning_rate": 2.1630179032755632e-06,
|
|
"loss": 0.7276,
|
|
"step": 646
|
|
},
|
|
{
|
|
"epoch": 0.701878093430063,
|
|
"grad_norm": 0.16039669148821492,
|
|
"learning_rate": 2.1485510093905264e-06,
|
|
"loss": 0.7245,
|
|
"step": 647
|
|
},
|
|
{
|
|
"epoch": 0.7029629127398468,
|
|
"grad_norm": 0.1702577848087486,
|
|
"learning_rate": 2.1341194063447533e-06,
|
|
"loss": 0.7264,
|
|
"step": 648
|
|
},
|
|
{
|
|
"epoch": 0.7040477320496304,
|
|
"grad_norm": 0.17234438323101783,
|
|
"learning_rate": 2.119723272750379e-06,
|
|
"loss": 0.7252,
|
|
"step": 649
|
|
},
|
|
{
|
|
"epoch": 0.7051325513594142,
|
|
"grad_norm": 0.16912223321743175,
|
|
"learning_rate": 2.105362786780547e-06,
|
|
"loss": 0.7229,
|
|
"step": 650
|
|
},
|
|
{
|
|
"epoch": 0.7051325513594142,
|
|
"eval_loss": 0.7264869809150696,
|
|
"eval_runtime": 108.6627,
|
|
"eval_samples_per_second": 87.739,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 650
|
|
},
|
|
{
|
|
"epoch": 0.7062173706691979,
|
|
"grad_norm": 0.1518545509592299,
|
|
"learning_rate": 2.0910381261672136e-06,
|
|
"loss": 0.7253,
|
|
"step": 651
|
|
},
|
|
{
|
|
"epoch": 0.7073021899789816,
|
|
"grad_norm": 0.16971215450252325,
|
|
"learning_rate": 2.076749468198943e-06,
|
|
"loss": 0.7267,
|
|
"step": 652
|
|
},
|
|
{
|
|
"epoch": 0.7083870092887653,
|
|
"grad_norm": 0.16219091114702947,
|
|
"learning_rate": 2.0624969897187084e-06,
|
|
"loss": 0.7252,
|
|
"step": 653
|
|
},
|
|
{
|
|
"epoch": 0.7094718285985491,
|
|
"grad_norm": 0.16197690320645672,
|
|
"learning_rate": 2.048280867121722e-06,
|
|
"loss": 0.7279,
|
|
"step": 654
|
|
},
|
|
{
|
|
"epoch": 0.7105566479083327,
|
|
"grad_norm": 0.15555593236095563,
|
|
"learning_rate": 2.0341012763532243e-06,
|
|
"loss": 0.7266,
|
|
"step": 655
|
|
},
|
|
{
|
|
"epoch": 0.7116414672181165,
|
|
"grad_norm": 0.17254981941367684,
|
|
"learning_rate": 2.019958392906332e-06,
|
|
"loss": 0.7298,
|
|
"step": 656
|
|
},
|
|
{
|
|
"epoch": 0.7127262865279002,
|
|
"grad_norm": 0.16803679989211034,
|
|
"learning_rate": 2.0058523918198473e-06,
|
|
"loss": 0.7233,
|
|
"step": 657
|
|
},
|
|
{
|
|
"epoch": 0.7138111058376839,
|
|
"grad_norm": 0.16099184424741314,
|
|
"learning_rate": 1.9917834476761037e-06,
|
|
"loss": 0.7283,
|
|
"step": 658
|
|
},
|
|
{
|
|
"epoch": 0.7148959251474676,
|
|
"grad_norm": 0.17600688627953956,
|
|
"learning_rate": 1.9777517345988057e-06,
|
|
"loss": 0.7279,
|
|
"step": 659
|
|
},
|
|
{
|
|
"epoch": 0.7159807444572514,
|
|
"grad_norm": 0.1599393258310897,
|
|
"learning_rate": 1.963757426250858e-06,
|
|
"loss": 0.7262,
|
|
"step": 660
|
|
},
|
|
{
|
|
"epoch": 0.7159807444572514,
|
|
"eval_loss": 0.7262735366821289,
|
|
"eval_runtime": 108.7906,
|
|
"eval_samples_per_second": 87.636,
|
|
"eval_steps_per_second": 0.689,
|
|
"step": 660
|
|
},
|
|
{
|
|
"epoch": 0.717065563767035,
|
|
"grad_norm": 0.16402767095037116,
|
|
"learning_rate": 1.949800695832236e-06,
|
|
"loss": 0.7256,
|
|
"step": 661
|
|
},
|
|
{
|
|
"epoch": 0.7181503830768188,
|
|
"grad_norm": 0.17637232160089736,
|
|
"learning_rate": 1.9358817160778272e-06,
|
|
"loss": 0.7291,
|
|
"step": 662
|
|
},
|
|
{
|
|
"epoch": 0.7192352023866024,
|
|
"grad_norm": 0.15091067397929198,
|
|
"learning_rate": 1.9220006592553075e-06,
|
|
"loss": 0.7243,
|
|
"step": 663
|
|
},
|
|
{
|
|
"epoch": 0.7203200216963862,
|
|
"grad_norm": 0.16329397517232444,
|
|
"learning_rate": 1.908157697162993e-06,
|
|
"loss": 0.7272,
|
|
"step": 664
|
|
},
|
|
{
|
|
"epoch": 0.7214048410061699,
|
|
"grad_norm": 0.17963032159288322,
|
|
"learning_rate": 1.8943530011277261e-06,
|
|
"loss": 0.7258,
|
|
"step": 665
|
|
},
|
|
{
|
|
"epoch": 0.7224896603159536,
|
|
"grad_norm": 0.1612233781915667,
|
|
"learning_rate": 1.8805867420027529e-06,
|
|
"loss": 0.7275,
|
|
"step": 666
|
|
},
|
|
{
|
|
"epoch": 0.7235744796257373,
|
|
"grad_norm": 0.16786914782825432,
|
|
"learning_rate": 1.866859090165598e-06,
|
|
"loss": 0.7287,
|
|
"step": 667
|
|
},
|
|
{
|
|
"epoch": 0.7246592989355211,
|
|
"grad_norm": 0.17518939037007217,
|
|
"learning_rate": 1.8531702155159792e-06,
|
|
"loss": 0.7232,
|
|
"step": 668
|
|
},
|
|
{
|
|
"epoch": 0.7257441182453047,
|
|
"grad_norm": 0.16119896959136812,
|
|
"learning_rate": 1.8395202874736752e-06,
|
|
"loss": 0.7277,
|
|
"step": 669
|
|
},
|
|
{
|
|
"epoch": 0.7268289375550885,
|
|
"grad_norm": 0.15926082041952538,
|
|
"learning_rate": 1.8259094749764532e-06,
|
|
"loss": 0.7256,
|
|
"step": 670
|
|
},
|
|
{
|
|
"epoch": 0.7268289375550885,
|
|
"eval_loss": 0.7260706424713135,
|
|
"eval_runtime": 108.7174,
|
|
"eval_samples_per_second": 87.695,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 670
|
|
},
|
|
{
|
|
"epoch": 0.7279137568648721,
|
|
"grad_norm": 0.16222434918580295,
|
|
"learning_rate": 1.8123379464779606e-06,
|
|
"loss": 0.7264,
|
|
"step": 671
|
|
},
|
|
{
|
|
"epoch": 0.7289985761746559,
|
|
"grad_norm": 0.1725638369615756,
|
|
"learning_rate": 1.7988058699456596e-06,
|
|
"loss": 0.7284,
|
|
"step": 672
|
|
},
|
|
{
|
|
"epoch": 0.7300833954844396,
|
|
"grad_norm": 0.1759812325291814,
|
|
"learning_rate": 1.7853134128587246e-06,
|
|
"loss": 0.7255,
|
|
"step": 673
|
|
},
|
|
{
|
|
"epoch": 0.7311682147942233,
|
|
"grad_norm": 0.1591812510571049,
|
|
"learning_rate": 1.771860742205988e-06,
|
|
"loss": 0.7253,
|
|
"step": 674
|
|
},
|
|
{
|
|
"epoch": 0.7322530341040071,
|
|
"grad_norm": 0.17928563081050367,
|
|
"learning_rate": 1.7584480244838687e-06,
|
|
"loss": 0.7252,
|
|
"step": 675
|
|
},
|
|
{
|
|
"epoch": 0.7333378534137908,
|
|
"grad_norm": 0.16871463749023005,
|
|
"learning_rate": 1.7450754256943014e-06,
|
|
"loss": 0.7271,
|
|
"step": 676
|
|
},
|
|
{
|
|
"epoch": 0.7344226727235745,
|
|
"grad_norm": 0.1788357380936577,
|
|
"learning_rate": 1.731743111342703e-06,
|
|
"loss": 0.7278,
|
|
"step": 677
|
|
},
|
|
{
|
|
"epoch": 0.7355074920333582,
|
|
"grad_norm": 0.17415287425610976,
|
|
"learning_rate": 1.7184512464358998e-06,
|
|
"loss": 0.7264,
|
|
"step": 678
|
|
},
|
|
{
|
|
"epoch": 0.736592311343142,
|
|
"grad_norm": 0.16409185847423846,
|
|
"learning_rate": 1.7051999954801058e-06,
|
|
"loss": 0.7264,
|
|
"step": 679
|
|
},
|
|
{
|
|
"epoch": 0.7376771306529256,
|
|
"grad_norm": 0.17148671830008877,
|
|
"learning_rate": 1.691989522478869e-06,
|
|
"loss": 0.724,
|
|
"step": 680
|
|
},
|
|
{
|
|
"epoch": 0.7376771306529256,
|
|
"eval_loss": 0.7258756160736084,
|
|
"eval_runtime": 108.6055,
|
|
"eval_samples_per_second": 87.786,
|
|
"eval_steps_per_second": 0.691,
|
|
"step": 680
|
|
},
|
|
{
|
|
"epoch": 0.7387619499627094,
|
|
"grad_norm": 0.1609146842916726,
|
|
"learning_rate": 1.6788199909310626e-06,
|
|
"loss": 0.7276,
|
|
"step": 681
|
|
},
|
|
{
|
|
"epoch": 0.739846769272493,
|
|
"grad_norm": 0.17169792032807274,
|
|
"learning_rate": 1.6656915638288423e-06,
|
|
"loss": 0.7287,
|
|
"step": 682
|
|
},
|
|
{
|
|
"epoch": 0.7409315885822768,
|
|
"grad_norm": 0.17450779056407184,
|
|
"learning_rate": 1.6526044036556349e-06,
|
|
"loss": 0.7315,
|
|
"step": 683
|
|
},
|
|
{
|
|
"epoch": 0.7420164078920605,
|
|
"grad_norm": 0.1648515137621003,
|
|
"learning_rate": 1.6395586723841328e-06,
|
|
"loss": 0.7247,
|
|
"step": 684
|
|
},
|
|
{
|
|
"epoch": 0.7431012272018442,
|
|
"grad_norm": 0.16744026588499902,
|
|
"learning_rate": 1.6265545314742838e-06,
|
|
"loss": 0.7268,
|
|
"step": 685
|
|
},
|
|
{
|
|
"epoch": 0.7441860465116279,
|
|
"grad_norm": 0.15958700866366168,
|
|
"learning_rate": 1.6135921418712959e-06,
|
|
"loss": 0.726,
|
|
"step": 686
|
|
},
|
|
{
|
|
"epoch": 0.7452708658214117,
|
|
"grad_norm": 0.1660679850836145,
|
|
"learning_rate": 1.600671664003639e-06,
|
|
"loss": 0.7241,
|
|
"step": 687
|
|
},
|
|
{
|
|
"epoch": 0.7463556851311953,
|
|
"grad_norm": 0.16390322586606818,
|
|
"learning_rate": 1.5877932577810712e-06,
|
|
"loss": 0.7218,
|
|
"step": 688
|
|
},
|
|
{
|
|
"epoch": 0.7474405044409791,
|
|
"grad_norm": 0.15636656647047326,
|
|
"learning_rate": 1.5749570825926437e-06,
|
|
"loss": 0.7273,
|
|
"step": 689
|
|
},
|
|
{
|
|
"epoch": 0.7485253237507627,
|
|
"grad_norm": 0.15413041269066685,
|
|
"learning_rate": 1.5621632973047468e-06,
|
|
"loss": 0.7241,
|
|
"step": 690
|
|
},
|
|
{
|
|
"epoch": 0.7485253237507627,
|
|
"eval_loss": 0.7256820201873779,
|
|
"eval_runtime": 108.6705,
|
|
"eval_samples_per_second": 87.733,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 690
|
|
},
|
|
{
|
|
"epoch": 0.7496101430605465,
|
|
"grad_norm": 0.15931337858369668,
|
|
"learning_rate": 1.5494120602591305e-06,
|
|
"loss": 0.7263,
|
|
"step": 691
|
|
},
|
|
{
|
|
"epoch": 0.7506949623703302,
|
|
"grad_norm": 0.16237019905864356,
|
|
"learning_rate": 1.5367035292709432e-06,
|
|
"loss": 0.7247,
|
|
"step": 692
|
|
},
|
|
{
|
|
"epoch": 0.751779781680114,
|
|
"grad_norm": 0.15237877763313265,
|
|
"learning_rate": 1.5240378616267887e-06,
|
|
"loss": 0.7276,
|
|
"step": 693
|
|
},
|
|
{
|
|
"epoch": 0.7528646009898976,
|
|
"grad_norm": 0.15604073580214003,
|
|
"learning_rate": 1.5114152140827744e-06,
|
|
"loss": 0.7261,
|
|
"step": 694
|
|
},
|
|
{
|
|
"epoch": 0.7539494202996814,
|
|
"grad_norm": 0.16190149615134655,
|
|
"learning_rate": 1.4988357428625711e-06,
|
|
"loss": 0.7261,
|
|
"step": 695
|
|
},
|
|
{
|
|
"epoch": 0.755034239609465,
|
|
"grad_norm": 0.16379175823219297,
|
|
"learning_rate": 1.4862996036554756e-06,
|
|
"loss": 0.725,
|
|
"step": 696
|
|
},
|
|
{
|
|
"epoch": 0.7561190589192488,
|
|
"grad_norm": 0.15009367703893553,
|
|
"learning_rate": 1.473806951614492e-06,
|
|
"loss": 0.7262,
|
|
"step": 697
|
|
},
|
|
{
|
|
"epoch": 0.7572038782290325,
|
|
"grad_norm": 0.15919308012136005,
|
|
"learning_rate": 1.4613579413544065e-06,
|
|
"loss": 0.7227,
|
|
"step": 698
|
|
},
|
|
{
|
|
"epoch": 0.7582886975388162,
|
|
"grad_norm": 0.1661228051986525,
|
|
"learning_rate": 1.4489527269498749e-06,
|
|
"loss": 0.7287,
|
|
"step": 699
|
|
},
|
|
{
|
|
"epoch": 0.7593735168485999,
|
|
"grad_norm": 0.15195694088930467,
|
|
"learning_rate": 1.4365914619335158e-06,
|
|
"loss": 0.7254,
|
|
"step": 700
|
|
},
|
|
{
|
|
"epoch": 0.7593735168485999,
|
|
"eval_loss": 0.7255001664161682,
|
|
"eval_runtime": 108.7009,
|
|
"eval_samples_per_second": 87.709,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 700
|
|
},
|
|
{
|
|
"epoch": 0.7604583361583837,
|
|
"grad_norm": 0.1607846909765903,
|
|
"learning_rate": 1.424274299294006e-06,
|
|
"loss": 0.7259,
|
|
"step": 701
|
|
},
|
|
{
|
|
"epoch": 0.7615431554681673,
|
|
"grad_norm": 0.15241158677973768,
|
|
"learning_rate": 1.412001391474196e-06,
|
|
"loss": 0.7222,
|
|
"step": 702
|
|
},
|
|
{
|
|
"epoch": 0.7626279747779511,
|
|
"grad_norm": 0.1493983763186097,
|
|
"learning_rate": 1.3997728903692164e-06,
|
|
"loss": 0.7292,
|
|
"step": 703
|
|
},
|
|
{
|
|
"epoch": 0.7637127940877347,
|
|
"grad_norm": 0.15475419901307835,
|
|
"learning_rate": 1.3875889473245996e-06,
|
|
"loss": 0.7286,
|
|
"step": 704
|
|
},
|
|
{
|
|
"epoch": 0.7647976133975185,
|
|
"grad_norm": 0.1569124690845288,
|
|
"learning_rate": 1.3754497131344097e-06,
|
|
"loss": 0.7287,
|
|
"step": 705
|
|
},
|
|
{
|
|
"epoch": 0.7658824327073022,
|
|
"grad_norm": 0.15709209563792223,
|
|
"learning_rate": 1.3633553380393677e-06,
|
|
"loss": 0.7278,
|
|
"step": 706
|
|
},
|
|
{
|
|
"epoch": 0.7669672520170859,
|
|
"grad_norm": 0.1517814869619425,
|
|
"learning_rate": 1.3513059717250037e-06,
|
|
"loss": 0.7277,
|
|
"step": 707
|
|
},
|
|
{
|
|
"epoch": 0.7680520713268696,
|
|
"grad_norm": 0.1513491806582122,
|
|
"learning_rate": 1.3393017633197958e-06,
|
|
"loss": 0.7297,
|
|
"step": 708
|
|
},
|
|
{
|
|
"epoch": 0.7691368906366534,
|
|
"grad_norm": 0.1520588299342006,
|
|
"learning_rate": 1.3273428613933298e-06,
|
|
"loss": 0.725,
|
|
"step": 709
|
|
},
|
|
{
|
|
"epoch": 0.770221709946437,
|
|
"grad_norm": 0.15383128405191898,
|
|
"learning_rate": 1.3154294139544516e-06,
|
|
"loss": 0.7251,
|
|
"step": 710
|
|
},
|
|
{
|
|
"epoch": 0.770221709946437,
|
|
"eval_loss": 0.7253365516662598,
|
|
"eval_runtime": 108.7066,
|
|
"eval_samples_per_second": 87.704,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 710
|
|
},
|
|
{
|
|
"epoch": 0.7713065292562208,
|
|
"grad_norm": 0.1563169647702222,
|
|
"learning_rate": 1.303561568449448e-06,
|
|
"loss": 0.7258,
|
|
"step": 711
|
|
},
|
|
{
|
|
"epoch": 0.7723913485660044,
|
|
"grad_norm": 0.15790530595831367,
|
|
"learning_rate": 1.2917394717602123e-06,
|
|
"loss": 0.7272,
|
|
"step": 712
|
|
},
|
|
{
|
|
"epoch": 0.7734761678757882,
|
|
"grad_norm": 0.15466165459446593,
|
|
"learning_rate": 1.2799632702024307e-06,
|
|
"loss": 0.724,
|
|
"step": 713
|
|
},
|
|
{
|
|
"epoch": 0.7745609871855719,
|
|
"grad_norm": 0.1476484964667729,
|
|
"learning_rate": 1.26823310952377e-06,
|
|
"loss": 0.7276,
|
|
"step": 714
|
|
},
|
|
{
|
|
"epoch": 0.7756458064953556,
|
|
"grad_norm": 0.15641965506252684,
|
|
"learning_rate": 1.256549134902072e-06,
|
|
"loss": 0.7255,
|
|
"step": 715
|
|
},
|
|
{
|
|
"epoch": 0.7767306258051393,
|
|
"grad_norm": 0.16566995237775228,
|
|
"learning_rate": 1.2449114909435611e-06,
|
|
"loss": 0.7273,
|
|
"step": 716
|
|
},
|
|
{
|
|
"epoch": 0.7778154451149231,
|
|
"grad_norm": 0.14937886311367543,
|
|
"learning_rate": 1.2333203216810514e-06,
|
|
"loss": 0.7224,
|
|
"step": 717
|
|
},
|
|
{
|
|
"epoch": 0.7789002644247067,
|
|
"grad_norm": 0.1484220962326093,
|
|
"learning_rate": 1.2217757705721662e-06,
|
|
"loss": 0.7264,
|
|
"step": 718
|
|
},
|
|
{
|
|
"epoch": 0.7799850837344905,
|
|
"grad_norm": 0.15052364605909016,
|
|
"learning_rate": 1.2102779804975574e-06,
|
|
"loss": 0.7238,
|
|
"step": 719
|
|
},
|
|
{
|
|
"epoch": 0.7810699030442741,
|
|
"grad_norm": 0.15353115572100884,
|
|
"learning_rate": 1.1988270937591446e-06,
|
|
"loss": 0.7253,
|
|
"step": 720
|
|
},
|
|
{
|
|
"epoch": 0.7810699030442741,
|
|
"eval_loss": 0.7251782417297363,
|
|
"eval_runtime": 108.804,
|
|
"eval_samples_per_second": 87.625,
|
|
"eval_steps_per_second": 0.689,
|
|
"step": 720
|
|
},
|
|
{
|
|
"epoch": 0.7821547223540579,
|
|
"grad_norm": 0.1512819973748618,
|
|
"learning_rate": 1.1874232520783486e-06,
|
|
"loss": 0.7236,
|
|
"step": 721
|
|
},
|
|
{
|
|
"epoch": 0.7832395416638416,
|
|
"grad_norm": 0.14770099165831163,
|
|
"learning_rate": 1.1760665965943402e-06,
|
|
"loss": 0.7252,
|
|
"step": 722
|
|
},
|
|
{
|
|
"epoch": 0.7843243609736253,
|
|
"grad_norm": 0.15438107662654746,
|
|
"learning_rate": 1.164757267862292e-06,
|
|
"loss": 0.7247,
|
|
"step": 723
|
|
},
|
|
{
|
|
"epoch": 0.785409180283409,
|
|
"grad_norm": 0.14573770717715565,
|
|
"learning_rate": 1.1534954058516357e-06,
|
|
"loss": 0.7283,
|
|
"step": 724
|
|
},
|
|
{
|
|
"epoch": 0.7864939995931928,
|
|
"grad_norm": 0.14885866264392547,
|
|
"learning_rate": 1.1422811499443375e-06,
|
|
"loss": 0.7239,
|
|
"step": 725
|
|
},
|
|
{
|
|
"epoch": 0.7875788189029764,
|
|
"grad_norm": 0.15129264932653552,
|
|
"learning_rate": 1.1311146389331667e-06,
|
|
"loss": 0.7215,
|
|
"step": 726
|
|
},
|
|
{
|
|
"epoch": 0.7886636382127602,
|
|
"grad_norm": 0.15832117832175482,
|
|
"learning_rate": 1.119996011019981e-06,
|
|
"loss": 0.7255,
|
|
"step": 727
|
|
},
|
|
{
|
|
"epoch": 0.7897484575225439,
|
|
"grad_norm": 0.15730074585140089,
|
|
"learning_rate": 1.1089254038140141e-06,
|
|
"loss": 0.7256,
|
|
"step": 728
|
|
},
|
|
{
|
|
"epoch": 0.7908332768323276,
|
|
"grad_norm": 0.1573902715061644,
|
|
"learning_rate": 1.0979029543301718e-06,
|
|
"loss": 0.7262,
|
|
"step": 729
|
|
},
|
|
{
|
|
"epoch": 0.7919180961421113,
|
|
"grad_norm": 0.16319845819543635,
|
|
"learning_rate": 1.0869287989873406e-06,
|
|
"loss": 0.7204,
|
|
"step": 730
|
|
},
|
|
{
|
|
"epoch": 0.7919180961421113,
|
|
"eval_loss": 0.7250498533248901,
|
|
"eval_runtime": 108.9615,
|
|
"eval_samples_per_second": 87.499,
|
|
"eval_steps_per_second": 0.688,
|
|
"step": 730
|
|
},
|
|
{
|
|
"epoch": 0.793002915451895,
|
|
"grad_norm": 0.14808854493903703,
|
|
"learning_rate": 1.0760030736066952e-06,
|
|
"loss": 0.7249,
|
|
"step": 731
|
|
},
|
|
{
|
|
"epoch": 0.7940877347616787,
|
|
"grad_norm": 0.15662829643507056,
|
|
"learning_rate": 1.0651259134100205e-06,
|
|
"loss": 0.7243,
|
|
"step": 732
|
|
},
|
|
{
|
|
"epoch": 0.7951725540714625,
|
|
"grad_norm": 0.15698257105935445,
|
|
"learning_rate": 1.0542974530180327e-06,
|
|
"loss": 0.7243,
|
|
"step": 733
|
|
},
|
|
{
|
|
"epoch": 0.7962573733812461,
|
|
"grad_norm": 0.15009825120582931,
|
|
"learning_rate": 1.0435178264487205e-06,
|
|
"loss": 0.7229,
|
|
"step": 734
|
|
},
|
|
{
|
|
"epoch": 0.7973421926910299,
|
|
"grad_norm": 0.16224102000760765,
|
|
"learning_rate": 1.0327871671156814e-06,
|
|
"loss": 0.7249,
|
|
"step": 735
|
|
},
|
|
{
|
|
"epoch": 0.7984270120008136,
|
|
"grad_norm": 0.14905409653469245,
|
|
"learning_rate": 1.022105607826473e-06,
|
|
"loss": 0.7248,
|
|
"step": 736
|
|
},
|
|
{
|
|
"epoch": 0.7995118313105973,
|
|
"grad_norm": 0.1465905416526197,
|
|
"learning_rate": 1.011473280780968e-06,
|
|
"loss": 0.7281,
|
|
"step": 737
|
|
},
|
|
{
|
|
"epoch": 0.800596650620381,
|
|
"grad_norm": 0.15900337010716586,
|
|
"learning_rate": 1.000890317569715e-06,
|
|
"loss": 0.7287,
|
|
"step": 738
|
|
},
|
|
{
|
|
"epoch": 0.8016814699301648,
|
|
"grad_norm": 0.15093512515655447,
|
|
"learning_rate": 9.903568491723176e-07,
|
|
"loss": 0.725,
|
|
"step": 739
|
|
},
|
|
{
|
|
"epoch": 0.8027662892399484,
|
|
"grad_norm": 0.1492907213057733,
|
|
"learning_rate": 9.798730059558076e-07,
|
|
"loss": 0.7239,
|
|
"step": 740
|
|
},
|
|
{
|
|
"epoch": 0.8027662892399484,
|
|
"eval_loss": 0.7249204516410828,
|
|
"eval_runtime": 108.8051,
|
|
"eval_samples_per_second": 87.625,
|
|
"eval_steps_per_second": 0.689,
|
|
"step": 740
|
|
},
|
|
{
|
|
"epoch": 0.8038511085497322,
|
|
"grad_norm": 0.1494555743822712,
|
|
"learning_rate": 9.69438917673033e-07,
|
|
"loss": 0.7263,
|
|
"step": 741
|
|
},
|
|
{
|
|
"epoch": 0.804935927859516,
|
|
"grad_norm": 0.14571069308734394,
|
|
"learning_rate": 9.59054713461049e-07,
|
|
"loss": 0.7238,
|
|
"step": 742
|
|
},
|
|
{
|
|
"epoch": 0.8060207471692996,
|
|
"grad_norm": 0.14964387496204115,
|
|
"learning_rate": 9.487205218395262e-07,
|
|
"loss": 0.7236,
|
|
"step": 743
|
|
},
|
|
{
|
|
"epoch": 0.8071055664790834,
|
|
"grad_norm": 0.14602726448076972,
|
|
"learning_rate": 9.384364707091559e-07,
|
|
"loss": 0.7237,
|
|
"step": 744
|
|
},
|
|
{
|
|
"epoch": 0.808190385788867,
|
|
"grad_norm": 0.15707406969931928,
|
|
"learning_rate": 9.282026873500666e-07,
|
|
"loss": 0.7242,
|
|
"step": 745
|
|
},
|
|
{
|
|
"epoch": 0.8092752050986508,
|
|
"grad_norm": 0.14711730382168384,
|
|
"learning_rate": 9.180192984202513e-07,
|
|
"loss": 0.7268,
|
|
"step": 746
|
|
},
|
|
{
|
|
"epoch": 0.8103600244084345,
|
|
"grad_norm": 0.1475210618723708,
|
|
"learning_rate": 9.078864299539963e-07,
|
|
"loss": 0.722,
|
|
"step": 747
|
|
},
|
|
{
|
|
"epoch": 0.8114448437182182,
|
|
"grad_norm": 0.1503680453933507,
|
|
"learning_rate": 8.978042073603243e-07,
|
|
"loss": 0.7286,
|
|
"step": 748
|
|
},
|
|
{
|
|
"epoch": 0.8125296630280019,
|
|
"grad_norm": 0.14551831472449375,
|
|
"learning_rate": 8.877727554214432e-07,
|
|
"loss": 0.7223,
|
|
"step": 749
|
|
},
|
|
{
|
|
"epoch": 0.8136144823377857,
|
|
"grad_norm": 0.14933248128984608,
|
|
"learning_rate": 8.777921982911996e-07,
|
|
"loss": 0.7244,
|
|
"step": 750
|
|
},
|
|
{
|
|
"epoch": 0.8136144823377857,
|
|
"eval_loss": 0.7247990965843201,
|
|
"eval_runtime": 108.6788,
|
|
"eval_samples_per_second": 87.726,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 750
|
|
},
|
|
{
|
|
"epoch": 0.8146993016475693,
|
|
"grad_norm": 0.150305371468892,
|
|
"learning_rate": 8.678626594935385e-07,
|
|
"loss": 0.7242,
|
|
"step": 751
|
|
},
|
|
{
|
|
"epoch": 0.8157841209573531,
|
|
"grad_norm": 0.15014173213213483,
|
|
"learning_rate": 8.579842619209844e-07,
|
|
"loss": 0.7231,
|
|
"step": 752
|
|
},
|
|
{
|
|
"epoch": 0.8168689402671367,
|
|
"grad_norm": 0.14676946111053624,
|
|
"learning_rate": 8.481571278331108e-07,
|
|
"loss": 0.7195,
|
|
"step": 753
|
|
},
|
|
{
|
|
"epoch": 0.8179537595769205,
|
|
"grad_norm": 0.14322775943055369,
|
|
"learning_rate": 8.383813788550326e-07,
|
|
"loss": 0.7222,
|
|
"step": 754
|
|
},
|
|
{
|
|
"epoch": 0.8190385788867042,
|
|
"grad_norm": 0.14578392504249468,
|
|
"learning_rate": 8.286571359758993e-07,
|
|
"loss": 0.7212,
|
|
"step": 755
|
|
},
|
|
{
|
|
"epoch": 0.8201233981964879,
|
|
"grad_norm": 0.14247573569748118,
|
|
"learning_rate": 8.18984519547395e-07,
|
|
"loss": 0.7218,
|
|
"step": 756
|
|
},
|
|
{
|
|
"epoch": 0.8212082175062716,
|
|
"grad_norm": 0.14659072439130408,
|
|
"learning_rate": 8.093636492822532e-07,
|
|
"loss": 0.727,
|
|
"step": 757
|
|
},
|
|
{
|
|
"epoch": 0.8222930368160554,
|
|
"grad_norm": 0.14954011587432003,
|
|
"learning_rate": 7.997946442527726e-07,
|
|
"loss": 0.7259,
|
|
"step": 758
|
|
},
|
|
{
|
|
"epoch": 0.823377856125839,
|
|
"grad_norm": 0.14576931031901047,
|
|
"learning_rate": 7.902776228893444e-07,
|
|
"loss": 0.7272,
|
|
"step": 759
|
|
},
|
|
{
|
|
"epoch": 0.8244626754356228,
|
|
"grad_norm": 0.1472554100775531,
|
|
"learning_rate": 7.808127029789869e-07,
|
|
"loss": 0.7261,
|
|
"step": 760
|
|
},
|
|
{
|
|
"epoch": 0.8244626754356228,
|
|
"eval_loss": 0.7246846556663513,
|
|
"eval_runtime": 108.8312,
|
|
"eval_samples_per_second": 87.604,
|
|
"eval_steps_per_second": 0.689,
|
|
"step": 760
|
|
},
|
|
{
|
|
"epoch": 0.8255474947454065,
|
|
"grad_norm": 0.1394428290196542,
|
|
"learning_rate": 7.714000016638829e-07,
|
|
"loss": 0.7221,
|
|
"step": 761
|
|
},
|
|
{
|
|
"epoch": 0.8266323140551902,
|
|
"grad_norm": 0.14870880577677073,
|
|
"learning_rate": 7.62039635439939e-07,
|
|
"loss": 0.723,
|
|
"step": 762
|
|
},
|
|
{
|
|
"epoch": 0.8277171333649739,
|
|
"grad_norm": 0.1531283031117994,
|
|
"learning_rate": 7.527317201553358e-07,
|
|
"loss": 0.7268,
|
|
"step": 763
|
|
},
|
|
{
|
|
"epoch": 0.8288019526747576,
|
|
"grad_norm": 0.14091906895489437,
|
|
"learning_rate": 7.434763710090991e-07,
|
|
"loss": 0.7267,
|
|
"step": 764
|
|
},
|
|
{
|
|
"epoch": 0.8298867719845413,
|
|
"grad_norm": 0.1407575155104757,
|
|
"learning_rate": 7.342737025496688e-07,
|
|
"loss": 0.725,
|
|
"step": 765
|
|
},
|
|
{
|
|
"epoch": 0.8309715912943251,
|
|
"grad_norm": 0.14429563419278052,
|
|
"learning_rate": 7.251238286734863e-07,
|
|
"loss": 0.7255,
|
|
"step": 766
|
|
},
|
|
{
|
|
"epoch": 0.8320564106041087,
|
|
"grad_norm": 0.146552949970586,
|
|
"learning_rate": 7.160268626235866e-07,
|
|
"loss": 0.7263,
|
|
"step": 767
|
|
},
|
|
{
|
|
"epoch": 0.8331412299138925,
|
|
"grad_norm": 0.14781604638174256,
|
|
"learning_rate": 7.06982916988187e-07,
|
|
"loss": 0.7255,
|
|
"step": 768
|
|
},
|
|
{
|
|
"epoch": 0.8342260492236762,
|
|
"grad_norm": 0.14172406706374732,
|
|
"learning_rate": 6.979921036993042e-07,
|
|
"loss": 0.7272,
|
|
"step": 769
|
|
},
|
|
{
|
|
"epoch": 0.8353108685334599,
|
|
"grad_norm": 0.14576979650977384,
|
|
"learning_rate": 6.890545340313609e-07,
|
|
"loss": 0.7261,
|
|
"step": 770
|
|
},
|
|
{
|
|
"epoch": 0.8353108685334599,
|
|
"eval_loss": 0.7245922088623047,
|
|
"eval_runtime": 108.7805,
|
|
"eval_samples_per_second": 87.644,
|
|
"eval_steps_per_second": 0.689,
|
|
"step": 770
|
|
},
|
|
{
|
|
"epoch": 0.8363956878432436,
|
|
"grad_norm": 0.15372865764730506,
|
|
"learning_rate": 6.801703185998165e-07,
|
|
"loss": 0.7258,
|
|
"step": 771
|
|
},
|
|
{
|
|
"epoch": 0.8374805071530274,
|
|
"grad_norm": 0.14076829292014756,
|
|
"learning_rate": 6.713395673597911e-07,
|
|
"loss": 0.7257,
|
|
"step": 772
|
|
},
|
|
{
|
|
"epoch": 0.838565326462811,
|
|
"grad_norm": 0.14645693346625896,
|
|
"learning_rate": 6.625623896047101e-07,
|
|
"loss": 0.7232,
|
|
"step": 773
|
|
},
|
|
{
|
|
"epoch": 0.8396501457725948,
|
|
"grad_norm": 0.14138216346190913,
|
|
"learning_rate": 6.538388939649442e-07,
|
|
"loss": 0.7215,
|
|
"step": 774
|
|
},
|
|
{
|
|
"epoch": 0.8407349650823784,
|
|
"grad_norm": 0.142370517145933,
|
|
"learning_rate": 6.451691884064737e-07,
|
|
"loss": 0.7234,
|
|
"step": 775
|
|
},
|
|
{
|
|
"epoch": 0.8418197843921622,
|
|
"grad_norm": 0.13948000581465836,
|
|
"learning_rate": 6.365533802295498e-07,
|
|
"loss": 0.7252,
|
|
"step": 776
|
|
},
|
|
{
|
|
"epoch": 0.8429046037019459,
|
|
"grad_norm": 0.14217547105914455,
|
|
"learning_rate": 6.279915760673593e-07,
|
|
"loss": 0.7267,
|
|
"step": 777
|
|
},
|
|
{
|
|
"epoch": 0.8439894230117296,
|
|
"grad_norm": 0.14615728937727987,
|
|
"learning_rate": 6.194838818847155e-07,
|
|
"loss": 0.7222,
|
|
"step": 778
|
|
},
|
|
{
|
|
"epoch": 0.8450742423215133,
|
|
"grad_norm": 0.13987405737502168,
|
|
"learning_rate": 6.110304029767372e-07,
|
|
"loss": 0.7253,
|
|
"step": 779
|
|
},
|
|
{
|
|
"epoch": 0.8461590616312971,
|
|
"grad_norm": 0.1444131083020489,
|
|
"learning_rate": 6.026312439675553e-07,
|
|
"loss": 0.7253,
|
|
"step": 780
|
|
},
|
|
{
|
|
"epoch": 0.8461590616312971,
|
|
"eval_loss": 0.7244901061058044,
|
|
"eval_runtime": 108.822,
|
|
"eval_samples_per_second": 87.611,
|
|
"eval_steps_per_second": 0.689,
|
|
"step": 780
|
|
},
|
|
{
|
|
"epoch": 0.8472438809410807,
|
|
"grad_norm": 0.14622984663099894,
|
|
"learning_rate": 5.942865088090088e-07,
|
|
"loss": 0.7261,
|
|
"step": 781
|
|
},
|
|
{
|
|
"epoch": 0.8483287002508645,
|
|
"grad_norm": 0.13775545945743656,
|
|
"learning_rate": 5.859963007793651e-07,
|
|
"loss": 0.7273,
|
|
"step": 782
|
|
},
|
|
{
|
|
"epoch": 0.8494135195606481,
|
|
"grad_norm": 0.1431958880765604,
|
|
"learning_rate": 5.77760722482037e-07,
|
|
"loss": 0.7256,
|
|
"step": 783
|
|
},
|
|
{
|
|
"epoch": 0.8504983388704319,
|
|
"grad_norm": 0.1447990850786756,
|
|
"learning_rate": 5.695798758443133e-07,
|
|
"loss": 0.7222,
|
|
"step": 784
|
|
},
|
|
{
|
|
"epoch": 0.8515831581802156,
|
|
"grad_norm": 0.15001921422654255,
|
|
"learning_rate": 5.614538621161036e-07,
|
|
"loss": 0.7242,
|
|
"step": 785
|
|
},
|
|
{
|
|
"epoch": 0.8526679774899993,
|
|
"grad_norm": 0.14389114694699778,
|
|
"learning_rate": 5.533827818686749e-07,
|
|
"loss": 0.7214,
|
|
"step": 786
|
|
},
|
|
{
|
|
"epoch": 0.853752796799783,
|
|
"grad_norm": 0.13888617179495713,
|
|
"learning_rate": 5.45366734993416e-07,
|
|
"loss": 0.7259,
|
|
"step": 787
|
|
},
|
|
{
|
|
"epoch": 0.8548376161095668,
|
|
"grad_norm": 0.13961527021408138,
|
|
"learning_rate": 5.374058207005945e-07,
|
|
"loss": 0.7262,
|
|
"step": 788
|
|
},
|
|
{
|
|
"epoch": 0.8559224354193504,
|
|
"grad_norm": 0.14408083207582856,
|
|
"learning_rate": 5.295001375181336e-07,
|
|
"loss": 0.7249,
|
|
"step": 789
|
|
},
|
|
{
|
|
"epoch": 0.8570072547291342,
|
|
"grad_norm": 0.1446548549097725,
|
|
"learning_rate": 5.216497832903927e-07,
|
|
"loss": 0.7266,
|
|
"step": 790
|
|
},
|
|
{
|
|
"epoch": 0.8570072547291342,
|
|
"eval_loss": 0.7244240045547485,
|
|
"eval_runtime": 108.6369,
|
|
"eval_samples_per_second": 87.76,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 790
|
|
},
|
|
{
|
|
"epoch": 0.8580920740389179,
|
|
"grad_norm": 0.14206827467632632,
|
|
"learning_rate": 5.138548551769512e-07,
|
|
"loss": 0.725,
|
|
"step": 791
|
|
},
|
|
{
|
|
"epoch": 0.8591768933487016,
|
|
"grad_norm": 0.13765523411051128,
|
|
"learning_rate": 5.061154496514125e-07,
|
|
"loss": 0.7274,
|
|
"step": 792
|
|
},
|
|
{
|
|
"epoch": 0.8602617126584853,
|
|
"grad_norm": 0.14461862018642865,
|
|
"learning_rate": 4.984316625002029e-07,
|
|
"loss": 0.7272,
|
|
"step": 793
|
|
},
|
|
{
|
|
"epoch": 0.861346531968269,
|
|
"grad_norm": 0.14159488968799147,
|
|
"learning_rate": 4.908035888213964e-07,
|
|
"loss": 0.7269,
|
|
"step": 794
|
|
},
|
|
{
|
|
"epoch": 0.8624313512780527,
|
|
"grad_norm": 0.1401730453066645,
|
|
"learning_rate": 4.832313230235253e-07,
|
|
"loss": 0.7287,
|
|
"step": 795
|
|
},
|
|
{
|
|
"epoch": 0.8635161705878365,
|
|
"grad_norm": 0.14430399475808467,
|
|
"learning_rate": 4.7571495882442363e-07,
|
|
"loss": 0.7241,
|
|
"step": 796
|
|
},
|
|
{
|
|
"epoch": 0.8646009898976201,
|
|
"grad_norm": 0.1434998185431347,
|
|
"learning_rate": 4.682545892500545e-07,
|
|
"loss": 0.7247,
|
|
"step": 797
|
|
},
|
|
{
|
|
"epoch": 0.8656858092074039,
|
|
"grad_norm": 0.14079724238639924,
|
|
"learning_rate": 4.608503066333742e-07,
|
|
"loss": 0.7281,
|
|
"step": 798
|
|
},
|
|
{
|
|
"epoch": 0.8667706285171876,
|
|
"grad_norm": 0.14166609606661967,
|
|
"learning_rate": 4.5350220261317633e-07,
|
|
"loss": 0.7234,
|
|
"step": 799
|
|
},
|
|
{
|
|
"epoch": 0.8678554478269713,
|
|
"grad_norm": 0.14209726292095484,
|
|
"learning_rate": 4.462103681329616e-07,
|
|
"loss": 0.721,
|
|
"step": 800
|
|
},
|
|
{
|
|
"epoch": 0.8678554478269713,
|
|
"eval_loss": 0.7243422269821167,
|
|
"eval_runtime": 108.5562,
|
|
"eval_samples_per_second": 87.825,
|
|
"eval_steps_per_second": 0.691,
|
|
"step": 800
|
|
},
|
|
{
|
|
"epoch": 0.868940267136755,
|
|
"grad_norm": 0.13859479140754644,
|
|
"learning_rate": 4.389748934398164e-07,
|
|
"loss": 0.724,
|
|
"step": 801
|
|
},
|
|
{
|
|
"epoch": 0.8700250864465388,
|
|
"grad_norm": 0.14139838653302136,
|
|
"learning_rate": 4.317958680832884e-07,
|
|
"loss": 0.7238,
|
|
"step": 802
|
|
},
|
|
{
|
|
"epoch": 0.8711099057563224,
|
|
"grad_norm": 0.13884780266816862,
|
|
"learning_rate": 4.246733809142889e-07,
|
|
"loss": 0.7247,
|
|
"step": 803
|
|
},
|
|
{
|
|
"epoch": 0.8721947250661062,
|
|
"grad_norm": 0.14158344919257543,
|
|
"learning_rate": 4.176075200839791e-07,
|
|
"loss": 0.7255,
|
|
"step": 804
|
|
},
|
|
{
|
|
"epoch": 0.8732795443758898,
|
|
"grad_norm": 0.1406274104017369,
|
|
"learning_rate": 4.105983730426916e-07,
|
|
"loss": 0.7249,
|
|
"step": 805
|
|
},
|
|
{
|
|
"epoch": 0.8743643636856736,
|
|
"grad_norm": 0.1436559977629329,
|
|
"learning_rate": 4.03646026538842e-07,
|
|
"loss": 0.7291,
|
|
"step": 806
|
|
},
|
|
{
|
|
"epoch": 0.8754491829954573,
|
|
"grad_norm": 0.13905977683367826,
|
|
"learning_rate": 3.9675056661785563e-07,
|
|
"loss": 0.7259,
|
|
"step": 807
|
|
},
|
|
{
|
|
"epoch": 0.876534002305241,
|
|
"grad_norm": 0.1350368671894585,
|
|
"learning_rate": 3.899120786211058e-07,
|
|
"loss": 0.7242,
|
|
"step": 808
|
|
},
|
|
{
|
|
"epoch": 0.8776188216150248,
|
|
"grad_norm": 0.14010167213253083,
|
|
"learning_rate": 3.8313064718485116e-07,
|
|
"loss": 0.7251,
|
|
"step": 809
|
|
},
|
|
{
|
|
"epoch": 0.8787036409248085,
|
|
"grad_norm": 0.14043441806084742,
|
|
"learning_rate": 3.7640635623919674e-07,
|
|
"loss": 0.7238,
|
|
"step": 810
|
|
},
|
|
{
|
|
"epoch": 0.8787036409248085,
|
|
"eval_loss": 0.7242914438247681,
|
|
"eval_runtime": 108.4291,
|
|
"eval_samples_per_second": 87.928,
|
|
"eval_steps_per_second": 0.692,
|
|
"step": 810
|
|
},
|
|
{
|
|
"epoch": 0.8797884602345922,
|
|
"grad_norm": 0.14597739195918666,
|
|
"learning_rate": 3.6973928900704503e-07,
|
|
"loss": 0.7223,
|
|
"step": 811
|
|
},
|
|
{
|
|
"epoch": 0.8808732795443759,
|
|
"grad_norm": 0.14004026532876637,
|
|
"learning_rate": 3.631295280030783e-07,
|
|
"loss": 0.7234,
|
|
"step": 812
|
|
},
|
|
{
|
|
"epoch": 0.8819580988541597,
|
|
"grad_norm": 0.13761772722113189,
|
|
"learning_rate": 3.5657715503272574e-07,
|
|
"loss": 0.7245,
|
|
"step": 813
|
|
},
|
|
{
|
|
"epoch": 0.8830429181639433,
|
|
"grad_norm": 0.13917740810160026,
|
|
"learning_rate": 3.500822511911578e-07,
|
|
"loss": 0.7257,
|
|
"step": 814
|
|
},
|
|
{
|
|
"epoch": 0.8841277374737271,
|
|
"grad_norm": 0.14077020735817927,
|
|
"learning_rate": 3.4364489686228076e-07,
|
|
"loss": 0.7272,
|
|
"step": 815
|
|
},
|
|
{
|
|
"epoch": 0.8852125567835107,
|
|
"grad_norm": 0.13849746520067135,
|
|
"learning_rate": 3.3726517171774163e-07,
|
|
"loss": 0.7228,
|
|
"step": 816
|
|
},
|
|
{
|
|
"epoch": 0.8862973760932945,
|
|
"grad_norm": 0.13945960915084693,
|
|
"learning_rate": 3.30943154715942e-07,
|
|
"loss": 0.7242,
|
|
"step": 817
|
|
},
|
|
{
|
|
"epoch": 0.8873821954030782,
|
|
"grad_norm": 0.14235867914664577,
|
|
"learning_rate": 3.2467892410106006e-07,
|
|
"loss": 0.7225,
|
|
"step": 818
|
|
},
|
|
{
|
|
"epoch": 0.8884670147128619,
|
|
"grad_norm": 0.14448235373129412,
|
|
"learning_rate": 3.1847255740208636e-07,
|
|
"loss": 0.7234,
|
|
"step": 819
|
|
},
|
|
{
|
|
"epoch": 0.8895518340226456,
|
|
"grad_norm": 0.13663699426090076,
|
|
"learning_rate": 3.1232413143185534e-07,
|
|
"loss": 0.7249,
|
|
"step": 820
|
|
},
|
|
{
|
|
"epoch": 0.8895518340226456,
|
|
"eval_loss": 0.7242299914360046,
|
|
"eval_runtime": 108.4679,
|
|
"eval_samples_per_second": 87.897,
|
|
"eval_steps_per_second": 0.691,
|
|
"step": 820
|
|
},
|
|
{
|
|
"epoch": 0.8906366533324294,
|
|
"grad_norm": 0.13908381651850316,
|
|
"learning_rate": 3.0623372228610725e-07,
|
|
"loss": 0.7252,
|
|
"step": 821
|
|
},
|
|
{
|
|
"epoch": 0.891721472642213,
|
|
"grad_norm": 0.13872995427912693,
|
|
"learning_rate": 3.0020140534253617e-07,
|
|
"loss": 0.7232,
|
|
"step": 822
|
|
},
|
|
{
|
|
"epoch": 0.8928062919519968,
|
|
"grad_norm": 0.13981744344012959,
|
|
"learning_rate": 2.942272552598596e-07,
|
|
"loss": 0.7256,
|
|
"step": 823
|
|
},
|
|
{
|
|
"epoch": 0.8938911112617804,
|
|
"grad_norm": 0.13968105284579016,
|
|
"learning_rate": 2.8831134597689604e-07,
|
|
"loss": 0.7211,
|
|
"step": 824
|
|
},
|
|
{
|
|
"epoch": 0.8949759305715642,
|
|
"grad_norm": 0.14232673834531245,
|
|
"learning_rate": 2.824537507116504e-07,
|
|
"loss": 0.7239,
|
|
"step": 825
|
|
},
|
|
{
|
|
"epoch": 0.8960607498813479,
|
|
"grad_norm": 0.14058100762514966,
|
|
"learning_rate": 2.7665454196040665e-07,
|
|
"loss": 0.7235,
|
|
"step": 826
|
|
},
|
|
{
|
|
"epoch": 0.8971455691911316,
|
|
"grad_norm": 0.1385223545477267,
|
|
"learning_rate": 2.7091379149682683e-07,
|
|
"loss": 0.7244,
|
|
"step": 827
|
|
},
|
|
{
|
|
"epoch": 0.8982303885009153,
|
|
"grad_norm": 0.1401420667592562,
|
|
"learning_rate": 2.652315703710712e-07,
|
|
"loss": 0.7255,
|
|
"step": 828
|
|
},
|
|
{
|
|
"epoch": 0.8993152078106991,
|
|
"grad_norm": 0.13765975035115535,
|
|
"learning_rate": 2.5960794890891093e-07,
|
|
"loss": 0.727,
|
|
"step": 829
|
|
},
|
|
{
|
|
"epoch": 0.9004000271204827,
|
|
"grad_norm": 0.1393723517733,
|
|
"learning_rate": 2.5404299671086264e-07,
|
|
"loss": 0.7254,
|
|
"step": 830
|
|
},
|
|
{
|
|
"epoch": 0.9004000271204827,
|
|
"eval_loss": 0.7241945266723633,
|
|
"eval_runtime": 108.6927,
|
|
"eval_samples_per_second": 87.715,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 830
|
|
},
|
|
{
|
|
"epoch": 0.9014848464302665,
|
|
"grad_norm": 0.136341269301018,
|
|
"learning_rate": 2.485367826513258e-07,
|
|
"loss": 0.7262,
|
|
"step": 831
|
|
},
|
|
{
|
|
"epoch": 0.9025696657400502,
|
|
"grad_norm": 0.13534524521877464,
|
|
"learning_rate": 2.4308937487772576e-07,
|
|
"loss": 0.7223,
|
|
"step": 832
|
|
},
|
|
{
|
|
"epoch": 0.9036544850498339,
|
|
"grad_norm": 0.1382147390177455,
|
|
"learning_rate": 2.3770084080967926e-07,
|
|
"loss": 0.725,
|
|
"step": 833
|
|
},
|
|
{
|
|
"epoch": 0.9047393043596176,
|
|
"grad_norm": 0.14138833642558557,
|
|
"learning_rate": 2.3237124713815285e-07,
|
|
"loss": 0.723,
|
|
"step": 834
|
|
},
|
|
{
|
|
"epoch": 0.9058241236694013,
|
|
"grad_norm": 0.13684451587905114,
|
|
"learning_rate": 2.2710065982464001e-07,
|
|
"loss": 0.7265,
|
|
"step": 835
|
|
},
|
|
{
|
|
"epoch": 0.906908942979185,
|
|
"grad_norm": 0.13632906600319858,
|
|
"learning_rate": 2.218891441003429e-07,
|
|
"loss": 0.7243,
|
|
"step": 836
|
|
},
|
|
{
|
|
"epoch": 0.9079937622889688,
|
|
"grad_norm": 0.13588777358371137,
|
|
"learning_rate": 2.1673676446536952e-07,
|
|
"loss": 0.7219,
|
|
"step": 837
|
|
},
|
|
{
|
|
"epoch": 0.9090785815987524,
|
|
"grad_norm": 0.13673010508746153,
|
|
"learning_rate": 2.1164358468793055e-07,
|
|
"loss": 0.725,
|
|
"step": 838
|
|
},
|
|
{
|
|
"epoch": 0.9101634009085362,
|
|
"grad_norm": 0.13533306331357944,
|
|
"learning_rate": 2.066096678035523e-07,
|
|
"loss": 0.7248,
|
|
"step": 839
|
|
},
|
|
{
|
|
"epoch": 0.9112482202183199,
|
|
"grad_norm": 0.1370765820418022,
|
|
"learning_rate": 2.0163507611429823e-07,
|
|
"loss": 0.723,
|
|
"step": 840
|
|
},
|
|
{
|
|
"epoch": 0.9112482202183199,
|
|
"eval_loss": 0.724153459072113,
|
|
"eval_runtime": 108.5965,
|
|
"eval_samples_per_second": 87.793,
|
|
"eval_steps_per_second": 0.691,
|
|
"step": 840
|
|
},
|
|
{
|
|
"epoch": 0.9123330395281036,
|
|
"grad_norm": 0.13542854082891065,
|
|
"learning_rate": 1.9671987118799307e-07,
|
|
"loss": 0.7247,
|
|
"step": 841
|
|
},
|
|
{
|
|
"epoch": 0.9134178588378873,
|
|
"grad_norm": 0.13790713518725453,
|
|
"learning_rate": 1.9186411385746507e-07,
|
|
"loss": 0.7239,
|
|
"step": 842
|
|
},
|
|
{
|
|
"epoch": 0.914502678147671,
|
|
"grad_norm": 0.13796552431922895,
|
|
"learning_rate": 1.870678642197926e-07,
|
|
"loss": 0.7219,
|
|
"step": 843
|
|
},
|
|
{
|
|
"epoch": 0.9155874974574547,
|
|
"grad_norm": 0.1375374114075323,
|
|
"learning_rate": 1.823311816355583e-07,
|
|
"loss": 0.726,
|
|
"step": 844
|
|
},
|
|
{
|
|
"epoch": 0.9166723167672385,
|
|
"grad_norm": 0.1377206455448937,
|
|
"learning_rate": 1.776541247281177e-07,
|
|
"loss": 0.7252,
|
|
"step": 845
|
|
},
|
|
{
|
|
"epoch": 0.9177571360770221,
|
|
"grad_norm": 0.1344192773735211,
|
|
"learning_rate": 1.730367513828679e-07,
|
|
"loss": 0.7232,
|
|
"step": 846
|
|
},
|
|
{
|
|
"epoch": 0.9188419553868059,
|
|
"grad_norm": 0.13376054584430533,
|
|
"learning_rate": 1.6847911874653843e-07,
|
|
"loss": 0.7235,
|
|
"step": 847
|
|
},
|
|
{
|
|
"epoch": 0.9199267746965896,
|
|
"grad_norm": 0.13439869668843996,
|
|
"learning_rate": 1.6398128322647865e-07,
|
|
"loss": 0.723,
|
|
"step": 848
|
|
},
|
|
{
|
|
"epoch": 0.9210115940063733,
|
|
"grad_norm": 0.13964459621458417,
|
|
"learning_rate": 1.5954330048996326e-07,
|
|
"loss": 0.7259,
|
|
"step": 849
|
|
},
|
|
{
|
|
"epoch": 0.922096413316157,
|
|
"grad_norm": 0.13629792697955317,
|
|
"learning_rate": 1.5516522546349833e-07,
|
|
"loss": 0.7197,
|
|
"step": 850
|
|
},
|
|
{
|
|
"epoch": 0.922096413316157,
|
|
"eval_loss": 0.7241156697273254,
|
|
"eval_runtime": 108.5644,
|
|
"eval_samples_per_second": 87.819,
|
|
"eval_steps_per_second": 0.691,
|
|
"step": 850
|
|
},
|
|
{
|
|
"epoch": 0.9231812326259408,
|
|
"grad_norm": 0.13493745494830567,
|
|
"learning_rate": 1.5084711233214699e-07,
|
|
"loss": 0.7243,
|
|
"step": 851
|
|
},
|
|
{
|
|
"epoch": 0.9242660519357244,
|
|
"grad_norm": 0.1356596248868148,
|
|
"learning_rate": 1.4658901453885654e-07,
|
|
"loss": 0.7234,
|
|
"step": 852
|
|
},
|
|
{
|
|
"epoch": 0.9253508712455082,
|
|
"grad_norm": 0.1389500426583187,
|
|
"learning_rate": 1.4239098478379565e-07,
|
|
"loss": 0.7252,
|
|
"step": 853
|
|
},
|
|
{
|
|
"epoch": 0.9264356905552918,
|
|
"grad_norm": 0.13346288169473994,
|
|
"learning_rate": 1.3825307502370487e-07,
|
|
"loss": 0.725,
|
|
"step": 854
|
|
},
|
|
{
|
|
"epoch": 0.9275205098650756,
|
|
"grad_norm": 0.1327056604810051,
|
|
"learning_rate": 1.3417533647125114e-07,
|
|
"loss": 0.7189,
|
|
"step": 855
|
|
},
|
|
{
|
|
"epoch": 0.9286053291748593,
|
|
"grad_norm": 0.13628540556310734,
|
|
"learning_rate": 1.3015781959439478e-07,
|
|
"loss": 0.7219,
|
|
"step": 856
|
|
},
|
|
{
|
|
"epoch": 0.929690148484643,
|
|
"grad_norm": 0.13501614840919324,
|
|
"learning_rate": 1.2620057411576568e-07,
|
|
"loss": 0.7259,
|
|
"step": 857
|
|
},
|
|
{
|
|
"epoch": 0.9307749677944267,
|
|
"grad_norm": 0.13881442153692194,
|
|
"learning_rate": 1.2230364901204773e-07,
|
|
"loss": 0.7223,
|
|
"step": 858
|
|
},
|
|
{
|
|
"epoch": 0.9318597871042105,
|
|
"grad_norm": 0.13320196272188387,
|
|
"learning_rate": 1.1846709251337129e-07,
|
|
"loss": 0.7222,
|
|
"step": 859
|
|
},
|
|
{
|
|
"epoch": 0.9329446064139941,
|
|
"grad_norm": 0.135540774249262,
|
|
"learning_rate": 1.1469095210271675e-07,
|
|
"loss": 0.7266,
|
|
"step": 860
|
|
},
|
|
{
|
|
"epoch": 0.9329446064139941,
|
|
"eval_loss": 0.7241001129150391,
|
|
"eval_runtime": 108.5809,
|
|
"eval_samples_per_second": 87.805,
|
|
"eval_steps_per_second": 0.691,
|
|
"step": 860
|
|
},
|
|
{
|
|
"epoch": 0.9340294257237779,
|
|
"grad_norm": 0.1380173343966705,
|
|
"learning_rate": 1.109752745153292e-07,
|
|
"loss": 0.7244,
|
|
"step": 861
|
|
},
|
|
{
|
|
"epoch": 0.9351142450335616,
|
|
"grad_norm": 0.13160062871740882,
|
|
"learning_rate": 1.0732010573813623e-07,
|
|
"loss": 0.7258,
|
|
"step": 862
|
|
},
|
|
{
|
|
"epoch": 0.9361990643433453,
|
|
"grad_norm": 0.13511481514710444,
|
|
"learning_rate": 1.0372549100918283e-07,
|
|
"loss": 0.7256,
|
|
"step": 863
|
|
},
|
|
{
|
|
"epoch": 0.937283883653129,
|
|
"grad_norm": 0.1314404956603575,
|
|
"learning_rate": 1.0019147481706626e-07,
|
|
"loss": 0.7212,
|
|
"step": 864
|
|
},
|
|
{
|
|
"epoch": 0.9383687029629127,
|
|
"grad_norm": 0.13703923561589082,
|
|
"learning_rate": 9.671810090039091e-08,
|
|
"loss": 0.7226,
|
|
"step": 865
|
|
},
|
|
{
|
|
"epoch": 0.9394535222726964,
|
|
"grad_norm": 0.13608577008694514,
|
|
"learning_rate": 9.330541224722378e-08,
|
|
"loss": 0.7231,
|
|
"step": 866
|
|
},
|
|
{
|
|
"epoch": 0.9405383415824802,
|
|
"grad_norm": 0.13806621521359003,
|
|
"learning_rate": 8.995345109456377e-08,
|
|
"loss": 0.7271,
|
|
"step": 867
|
|
},
|
|
{
|
|
"epoch": 0.9416231608922638,
|
|
"grad_norm": 0.1353062029172663,
|
|
"learning_rate": 8.666225892781765e-08,
|
|
"loss": 0.7258,
|
|
"step": 868
|
|
},
|
|
{
|
|
"epoch": 0.9427079802020476,
|
|
"grad_norm": 0.13680617294705663,
|
|
"learning_rate": 8.343187648028772e-08,
|
|
"loss": 0.7217,
|
|
"step": 869
|
|
},
|
|
{
|
|
"epoch": 0.9437927995118313,
|
|
"grad_norm": 0.13493382983403393,
|
|
"learning_rate": 8.026234373266773e-08,
|
|
"loss": 0.7247,
|
|
"step": 870
|
|
},
|
|
{
|
|
"epoch": 0.9437927995118313,
|
|
"eval_loss": 0.724071741104126,
|
|
"eval_runtime": 108.7638,
|
|
"eval_samples_per_second": 87.658,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 870
|
|
},
|
|
{
|
|
"epoch": 0.944877618821615,
|
|
"grad_norm": 0.13448724665345346,
|
|
"learning_rate": 7.715369991254662e-08,
|
|
"loss": 0.7244,
|
|
"step": 871
|
|
},
|
|
{
|
|
"epoch": 0.9459624381313987,
|
|
"grad_norm": 0.1356028773742455,
|
|
"learning_rate": 7.410598349392506e-08,
|
|
"loss": 0.724,
|
|
"step": 872
|
|
},
|
|
{
|
|
"epoch": 0.9470472574411825,
|
|
"grad_norm": 0.13612537206569597,
|
|
"learning_rate": 7.1119232196738e-08,
|
|
"loss": 0.7274,
|
|
"step": 873
|
|
},
|
|
{
|
|
"epoch": 0.9481320767509661,
|
|
"grad_norm": 0.13883570821688923,
|
|
"learning_rate": 6.819348298638839e-08,
|
|
"loss": 0.7263,
|
|
"step": 874
|
|
},
|
|
{
|
|
"epoch": 0.9492168960607499,
|
|
"grad_norm": 0.13264265764048075,
|
|
"learning_rate": 6.532877207328813e-08,
|
|
"loss": 0.7232,
|
|
"step": 875
|
|
},
|
|
{
|
|
"epoch": 0.9503017153705337,
|
|
"grad_norm": 0.13954361495637418,
|
|
"learning_rate": 6.252513491241285e-08,
|
|
"loss": 0.7234,
|
|
"step": 876
|
|
},
|
|
{
|
|
"epoch": 0.9513865346803173,
|
|
"grad_norm": 0.13360197271638555,
|
|
"learning_rate": 5.978260620286058e-08,
|
|
"loss": 0.723,
|
|
"step": 877
|
|
},
|
|
{
|
|
"epoch": 0.9524713539901011,
|
|
"grad_norm": 0.1314637699564161,
|
|
"learning_rate": 5.7101219887423233e-08,
|
|
"loss": 0.7257,
|
|
"step": 878
|
|
},
|
|
{
|
|
"epoch": 0.9535561732998847,
|
|
"grad_norm": 0.13392106802513146,
|
|
"learning_rate": 5.448100915216636e-08,
|
|
"loss": 0.7232,
|
|
"step": 879
|
|
},
|
|
{
|
|
"epoch": 0.9546409926096685,
|
|
"grad_norm": 0.13522662007382738,
|
|
"learning_rate": 5.192200642601841e-08,
|
|
"loss": 0.7224,
|
|
"step": 880
|
|
},
|
|
{
|
|
"epoch": 0.9546409926096685,
|
|
"eval_loss": 0.7240606546401978,
|
|
"eval_runtime": 108.7064,
|
|
"eval_samples_per_second": 87.704,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 880
|
|
},
|
|
{
|
|
"epoch": 0.9557258119194522,
|
|
"grad_norm": 0.13428639012820798,
|
|
"learning_rate": 4.942424338037044e-08,
|
|
"loss": 0.7241,
|
|
"step": 881
|
|
},
|
|
{
|
|
"epoch": 0.9568106312292359,
|
|
"grad_norm": 0.135647942541001,
|
|
"learning_rate": 4.6987750928682017e-08,
|
|
"loss": 0.7289,
|
|
"step": 882
|
|
},
|
|
{
|
|
"epoch": 0.9578954505390196,
|
|
"grad_norm": 0.13348778511819145,
|
|
"learning_rate": 4.461255922609986e-08,
|
|
"loss": 0.7258,
|
|
"step": 883
|
|
},
|
|
{
|
|
"epoch": 0.9589802698488034,
|
|
"grad_norm": 0.13524480850309303,
|
|
"learning_rate": 4.2298697669084785e-08,
|
|
"loss": 0.7226,
|
|
"step": 884
|
|
},
|
|
{
|
|
"epoch": 0.960065089158587,
|
|
"grad_norm": 0.1342302442153707,
|
|
"learning_rate": 4.004619489504813e-08,
|
|
"loss": 0.7244,
|
|
"step": 885
|
|
},
|
|
{
|
|
"epoch": 0.9611499084683708,
|
|
"grad_norm": 0.1359282017879408,
|
|
"learning_rate": 3.78550787819959e-08,
|
|
"loss": 0.726,
|
|
"step": 886
|
|
},
|
|
{
|
|
"epoch": 0.9622347277781544,
|
|
"grad_norm": 0.13610996574817735,
|
|
"learning_rate": 3.5725376448185744e-08,
|
|
"loss": 0.7219,
|
|
"step": 887
|
|
},
|
|
{
|
|
"epoch": 0.9633195470879382,
|
|
"grad_norm": 0.13224949770906974,
|
|
"learning_rate": 3.365711425178886e-08,
|
|
"loss": 0.7214,
|
|
"step": 888
|
|
},
|
|
{
|
|
"epoch": 0.9644043663977219,
|
|
"grad_norm": 0.13513692101369765,
|
|
"learning_rate": 3.165031779056582e-08,
|
|
"loss": 0.7261,
|
|
"step": 889
|
|
},
|
|
{
|
|
"epoch": 0.9654891857075056,
|
|
"grad_norm": 0.13462719071057394,
|
|
"learning_rate": 2.970501190154962e-08,
|
|
"loss": 0.7234,
|
|
"step": 890
|
|
},
|
|
{
|
|
"epoch": 0.9654891857075056,
|
|
"eval_loss": 0.7240492105484009,
|
|
"eval_runtime": 108.5574,
|
|
"eval_samples_per_second": 87.825,
|
|
"eval_steps_per_second": 0.691,
|
|
"step": 890
|
|
},
|
|
{
|
|
"epoch": 0.9665740050172893,
|
|
"grad_norm": 0.13271868273567594,
|
|
"learning_rate": 2.782122066073756e-08,
|
|
"loss": 0.7244,
|
|
"step": 891
|
|
},
|
|
{
|
|
"epoch": 0.9676588243270731,
|
|
"grad_norm": 0.13708737788809622,
|
|
"learning_rate": 2.5998967382792618e-08,
|
|
"loss": 0.7228,
|
|
"step": 892
|
|
},
|
|
{
|
|
"epoch": 0.9687436436368567,
|
|
"grad_norm": 0.13531886494212692,
|
|
"learning_rate": 2.423827462075701e-08,
|
|
"loss": 0.7279,
|
|
"step": 893
|
|
},
|
|
{
|
|
"epoch": 0.9698284629466405,
|
|
"grad_norm": 0.1359738222660984,
|
|
"learning_rate": 2.2539164165770178e-08,
|
|
"loss": 0.727,
|
|
"step": 894
|
|
},
|
|
{
|
|
"epoch": 0.9709132822564241,
|
|
"grad_norm": 0.13644782440998746,
|
|
"learning_rate": 2.090165704680236e-08,
|
|
"loss": 0.7232,
|
|
"step": 895
|
|
},
|
|
{
|
|
"epoch": 0.9719981015662079,
|
|
"grad_norm": 0.13486213998423852,
|
|
"learning_rate": 1.9325773530391446e-08,
|
|
"loss": 0.7257,
|
|
"step": 896
|
|
},
|
|
{
|
|
"epoch": 0.9730829208759916,
|
|
"grad_norm": 0.13705541543187244,
|
|
"learning_rate": 1.7811533120394296e-08,
|
|
"loss": 0.724,
|
|
"step": 897
|
|
},
|
|
{
|
|
"epoch": 0.9741677401857753,
|
|
"grad_norm": 0.1371546924335268,
|
|
"learning_rate": 1.6358954557744166e-08,
|
|
"loss": 0.7221,
|
|
"step": 898
|
|
},
|
|
{
|
|
"epoch": 0.975252559495559,
|
|
"grad_norm": 0.13337146572350866,
|
|
"learning_rate": 1.4968055820218653e-08,
|
|
"loss": 0.7223,
|
|
"step": 899
|
|
},
|
|
{
|
|
"epoch": 0.9763373788053428,
|
|
"grad_norm": 0.1312830829036625,
|
|
"learning_rate": 1.3638854122218214e-08,
|
|
"loss": 0.7271,
|
|
"step": 900
|
|
},
|
|
{
|
|
"epoch": 0.9763373788053428,
|
|
"eval_loss": 0.7240499258041382,
|
|
"eval_runtime": 108.4546,
|
|
"eval_samples_per_second": 87.908,
|
|
"eval_steps_per_second": 0.692,
|
|
"step": 900
|
|
},
|
|
{
|
|
"epoch": 0.9774221981151264,
|
|
"grad_norm": 0.13277090681049394,
|
|
"learning_rate": 1.2371365914551903e-08,
|
|
"loss": 0.7248,
|
|
"step": 901
|
|
},
|
|
{
|
|
"epoch": 0.9785070174249102,
|
|
"grad_norm": 0.13499989591222544,
|
|
"learning_rate": 1.1165606884234182e-08,
|
|
"loss": 0.7267,
|
|
"step": 902
|
|
},
|
|
{
|
|
"epoch": 0.9795918367346939,
|
|
"grad_norm": 0.13317601577729554,
|
|
"learning_rate": 1.0021591954291754e-08,
|
|
"loss": 0.7213,
|
|
"step": 903
|
|
},
|
|
{
|
|
"epoch": 0.9806766560444776,
|
|
"grad_norm": 0.1335874055337735,
|
|
"learning_rate": 8.939335283577599e-09,
|
|
"loss": 0.7232,
|
|
"step": 904
|
|
},
|
|
{
|
|
"epoch": 0.9817614753542613,
|
|
"grad_norm": 0.13717545379093252,
|
|
"learning_rate": 7.918850266596112e-09,
|
|
"loss": 0.7263,
|
|
"step": 905
|
|
},
|
|
{
|
|
"epoch": 0.982846294664045,
|
|
"grad_norm": 0.13178525458547813,
|
|
"learning_rate": 6.960149533337124e-09,
|
|
"loss": 0.7204,
|
|
"step": 906
|
|
},
|
|
{
|
|
"epoch": 0.9839311139738287,
|
|
"grad_norm": 0.13293535543691803,
|
|
"learning_rate": 6.063244949120473e-09,
|
|
"loss": 0.7238,
|
|
"step": 907
|
|
},
|
|
{
|
|
"epoch": 0.9850159332836125,
|
|
"grad_norm": 0.13294400889862482,
|
|
"learning_rate": 5.228147614448342e-09,
|
|
"loss": 0.7212,
|
|
"step": 908
|
|
},
|
|
{
|
|
"epoch": 0.9861007525933961,
|
|
"grad_norm": 0.1320390651519463,
|
|
"learning_rate": 4.4548678648681506e-09,
|
|
"loss": 0.7235,
|
|
"step": 909
|
|
},
|
|
{
|
|
"epoch": 0.9871855719031799,
|
|
"grad_norm": 0.13297478403106067,
|
|
"learning_rate": 3.7434152708437645e-09,
|
|
"loss": 0.7233,
|
|
"step": 910
|
|
},
|
|
{
|
|
"epoch": 0.9871855719031799,
|
|
"eval_loss": 0.7240437269210815,
|
|
"eval_runtime": 108.7272,
|
|
"eval_samples_per_second": 87.687,
|
|
"eval_steps_per_second": 0.69,
|
|
"step": 910
|
|
},
|
|
{
|
|
"epoch": 0.9882703912129636,
|
|
"grad_norm": 0.13353370639789847,
|
|
"learning_rate": 3.093798637638923e-09,
|
|
"loss": 0.726,
|
|
"step": 911
|
|
},
|
|
{
|
|
"epoch": 0.9893552105227473,
|
|
"grad_norm": 0.13350248025260392,
|
|
"learning_rate": 2.5060260052067742e-09,
|
|
"loss": 0.7227,
|
|
"step": 912
|
|
},
|
|
{
|
|
"epoch": 0.990440029832531,
|
|
"grad_norm": 0.13627772859668344,
|
|
"learning_rate": 1.980104648090508e-09,
|
|
"loss": 0.725,
|
|
"step": 913
|
|
},
|
|
{
|
|
"epoch": 0.9915248491423148,
|
|
"grad_norm": 0.1329148451322538,
|
|
"learning_rate": 1.51604107533454e-09,
|
|
"loss": 0.7205,
|
|
"step": 914
|
|
},
|
|
{
|
|
"epoch": 0.9926096684520984,
|
|
"grad_norm": 0.13161405037237076,
|
|
"learning_rate": 1.1138410304012415e-09,
|
|
"loss": 0.7234,
|
|
"step": 915
|
|
},
|
|
{
|
|
"epoch": 0.9936944877618822,
|
|
"grad_norm": 0.1347730166694005,
|
|
"learning_rate": 7.735094911032193e-10,
|
|
"loss": 0.7263,
|
|
"step": 916
|
|
},
|
|
{
|
|
"epoch": 0.9947793070716658,
|
|
"grad_norm": 0.1375437823642148,
|
|
"learning_rate": 4.950506695394763e-10,
|
|
"loss": 0.7274,
|
|
"step": 917
|
|
},
|
|
{
|
|
"epoch": 0.9958641263814496,
|
|
"grad_norm": 0.13354422205504643,
|
|
"learning_rate": 2.784680120437866e-10,
|
|
"loss": 0.7273,
|
|
"step": 918
|
|
},
|
|
{
|
|
"epoch": 0.9969489456912333,
|
|
"grad_norm": 0.13353910809068179,
|
|
"learning_rate": 1.237641991425065e-10,
|
|
"loss": 0.7237,
|
|
"step": 919
|
|
},
|
|
{
|
|
"epoch": 0.998033765001017,
|
|
"grad_norm": 0.13442440304803116,
|
|
"learning_rate": 3.094114552126826e-11,
|
|
"loss": 0.7253,
|
|
"step": 920
|
|
},
|
|
{
|
|
"epoch": 0.998033765001017,
|
|
"eval_loss": 0.7240483164787292,
|
|
"eval_runtime": 108.866,
|
|
"eval_samples_per_second": 87.576,
|
|
"eval_steps_per_second": 0.689,
|
|
"step": 920
|
|
},
|
|
{
|
|
"epoch": 0.9991185843108007,
|
|
"grad_norm": 0.13608136520076863,
|
|
"learning_rate": 0.0,
|
|
"loss": 0.7251,
|
|
"step": 921
|
|
},
|
|
{
|
|
"epoch": 0.9991185843108007,
|
|
"step": 921,
|
|
"total_flos": 1.23408369057792e+16,
|
|
"train_loss": 0.752030895211927,
|
|
"train_runtime": 60917.9213,
|
|
"train_samples_per_second": 15.495,
|
|
"train_steps_per_second": 0.015
|
|
}
|
|
],
|
|
"logging_steps": 1.0,
|
|
"max_steps": 921,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": true
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1.23408369057792e+16,
|
|
"train_batch_size": 1,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|