1956 lines
53 KiB
JSON
1956 lines
53 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 1.0,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 1910,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"entropy": 1.4903042420744896,
|
||
|
|
"epoch": 0.005235944760782774,
|
||
|
|
"grad_norm": 7.59375,
|
||
|
|
"learning_rate": 1.1780104712041885e-06,
|
||
|
|
"loss": 1.6362,
|
||
|
|
"mean_token_accuracy": 0.6602190021425486,
|
||
|
|
"num_tokens": 148667.0,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.5531679213047027,
|
||
|
|
"epoch": 0.010471889521565548,
|
||
|
|
"grad_norm": 6.9375,
|
||
|
|
"learning_rate": 2.486910994764398e-06,
|
||
|
|
"loss": 1.7071,
|
||
|
|
"mean_token_accuracy": 0.6444286152720451,
|
||
|
|
"num_tokens": 289380.0,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.5901738341897727,
|
||
|
|
"epoch": 0.015707834282348322,
|
||
|
|
"grad_norm": 5.65625,
|
||
|
|
"learning_rate": 3.7958115183246074e-06,
|
||
|
|
"loss": 1.6751,
|
||
|
|
"mean_token_accuracy": 0.6434450890868902,
|
||
|
|
"num_tokens": 433015.0,
|
||
|
|
"step": 30
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.5159442014992237,
|
||
|
|
"epoch": 0.020943779043131095,
|
||
|
|
"grad_norm": 5.0625,
|
||
|
|
"learning_rate": 5.104712041884817e-06,
|
||
|
|
"loss": 1.5104,
|
||
|
|
"mean_token_accuracy": 0.6608107829466462,
|
||
|
|
"num_tokens": 586196.0,
|
||
|
|
"step": 40
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.5390321850776671,
|
||
|
|
"epoch": 0.026179723803913868,
|
||
|
|
"grad_norm": 3.0,
|
||
|
|
"learning_rate": 6.4136125654450265e-06,
|
||
|
|
"loss": 1.4851,
|
||
|
|
"mean_token_accuracy": 0.664670011587441,
|
||
|
|
"num_tokens": 732500.0,
|
||
|
|
"step": 50
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.486197516322136,
|
||
|
|
"epoch": 0.031415668564696644,
|
||
|
|
"grad_norm": 2.703125,
|
||
|
|
"learning_rate": 7.722513089005236e-06,
|
||
|
|
"loss": 1.4375,
|
||
|
|
"mean_token_accuracy": 0.6728810863569379,
|
||
|
|
"num_tokens": 871796.0,
|
||
|
|
"step": 60
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.4761194687336683,
|
||
|
|
"epoch": 0.036651613325479414,
|
||
|
|
"grad_norm": 2.265625,
|
||
|
|
"learning_rate": 9.031413612565446e-06,
|
||
|
|
"loss": 1.4175,
|
||
|
|
"mean_token_accuracy": 0.6700865641236305,
|
||
|
|
"num_tokens": 1015760.0,
|
||
|
|
"step": 70
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.397418873384595,
|
||
|
|
"epoch": 0.04188755808626219,
|
||
|
|
"grad_norm": 2.25,
|
||
|
|
"learning_rate": 1.0340314136125655e-05,
|
||
|
|
"loss": 1.3627,
|
||
|
|
"mean_token_accuracy": 0.6793193189427257,
|
||
|
|
"num_tokens": 1162521.0,
|
||
|
|
"step": 80
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.4198294993489982,
|
||
|
|
"epoch": 0.04712350284704497,
|
||
|
|
"grad_norm": 2.25,
|
||
|
|
"learning_rate": 1.1649214659685865e-05,
|
||
|
|
"loss": 1.3982,
|
||
|
|
"mean_token_accuracy": 0.678496933169663,
|
||
|
|
"num_tokens": 1304606.0,
|
||
|
|
"step": 90
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.3636814955621959,
|
||
|
|
"epoch": 0.052359447607827736,
|
||
|
|
"grad_norm": 1.9140625,
|
||
|
|
"learning_rate": 1.2958115183246074e-05,
|
||
|
|
"loss": 1.3157,
|
||
|
|
"mean_token_accuracy": 0.6834567856043577,
|
||
|
|
"num_tokens": 1459672.0,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.371455504372716,
|
||
|
|
"epoch": 0.05759539236861051,
|
||
|
|
"grad_norm": 2.578125,
|
||
|
|
"learning_rate": 1.4267015706806284e-05,
|
||
|
|
"loss": 1.337,
|
||
|
|
"mean_token_accuracy": 0.680043394304812,
|
||
|
|
"num_tokens": 1614513.0,
|
||
|
|
"step": 110
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.4191205739974975,
|
||
|
|
"epoch": 0.06283133712939329,
|
||
|
|
"grad_norm": 2.4375,
|
||
|
|
"learning_rate": 1.5575916230366495e-05,
|
||
|
|
"loss": 1.3601,
|
||
|
|
"mean_token_accuracy": 0.6790978884324431,
|
||
|
|
"num_tokens": 1764120.0,
|
||
|
|
"step": 120
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.4203862166032195,
|
||
|
|
"epoch": 0.06806728189017607,
|
||
|
|
"grad_norm": 2.421875,
|
||
|
|
"learning_rate": 1.68848167539267e-05,
|
||
|
|
"loss": 1.3384,
|
||
|
|
"mean_token_accuracy": 0.6819259503856301,
|
||
|
|
"num_tokens": 1905542.0,
|
||
|
|
"step": 130
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.4149893302470447,
|
||
|
|
"epoch": 0.07330322665095883,
|
||
|
|
"grad_norm": 2.484375,
|
||
|
|
"learning_rate": 1.8193717277486914e-05,
|
||
|
|
"loss": 1.3124,
|
||
|
|
"mean_token_accuracy": 0.6826399000361562,
|
||
|
|
"num_tokens": 2051546.0,
|
||
|
|
"step": 140
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.319907895848155,
|
||
|
|
"epoch": 0.0785391714117416,
|
||
|
|
"grad_norm": 2.046875,
|
||
|
|
"learning_rate": 1.950261780104712e-05,
|
||
|
|
"loss": 1.2291,
|
||
|
|
"mean_token_accuracy": 0.6927984276786446,
|
||
|
|
"num_tokens": 2197518.0,
|
||
|
|
"step": 150
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.3174231912940741,
|
||
|
|
"epoch": 0.08377511617252438,
|
||
|
|
"grad_norm": 1.9453125,
|
||
|
|
"learning_rate": 2.0811518324607333e-05,
|
||
|
|
"loss": 1.2863,
|
||
|
|
"mean_token_accuracy": 0.6850969936698675,
|
||
|
|
"num_tokens": 2348210.0,
|
||
|
|
"step": 160
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2859620593488217,
|
||
|
|
"epoch": 0.08901106093330716,
|
||
|
|
"grad_norm": 2.5,
|
||
|
|
"learning_rate": 2.212041884816754e-05,
|
||
|
|
"loss": 1.2469,
|
||
|
|
"mean_token_accuracy": 0.694661033526063,
|
||
|
|
"num_tokens": 2488974.0,
|
||
|
|
"step": 170
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.3225422732532024,
|
||
|
|
"epoch": 0.09424700569408993,
|
||
|
|
"grad_norm": 2.203125,
|
||
|
|
"learning_rate": 2.3429319371727752e-05,
|
||
|
|
"loss": 1.3216,
|
||
|
|
"mean_token_accuracy": 0.6867891995236277,
|
||
|
|
"num_tokens": 2638402.0,
|
||
|
|
"step": 180
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2810360241681338,
|
||
|
|
"epoch": 0.0994829504548727,
|
||
|
|
"grad_norm": 2.09375,
|
||
|
|
"learning_rate": 2.473821989528796e-05,
|
||
|
|
"loss": 1.2682,
|
||
|
|
"mean_token_accuracy": 0.6961364889517426,
|
||
|
|
"num_tokens": 2780719.0,
|
||
|
|
"step": 190
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.30760794095695,
|
||
|
|
"epoch": 0.10471889521565547,
|
||
|
|
"grad_norm": 4.0625,
|
||
|
|
"learning_rate": 2.604712041884817e-05,
|
||
|
|
"loss": 1.3035,
|
||
|
|
"mean_token_accuracy": 0.6898889668285847,
|
||
|
|
"num_tokens": 2926604.0,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.264478771481663,
|
||
|
|
"epoch": 0.10995483997643825,
|
||
|
|
"grad_norm": 1.90625,
|
||
|
|
"learning_rate": 2.7356020942408378e-05,
|
||
|
|
"loss": 1.2493,
|
||
|
|
"mean_token_accuracy": 0.6941331747919322,
|
||
|
|
"num_tokens": 3070548.0,
|
||
|
|
"step": 210
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2868385933339597,
|
||
|
|
"epoch": 0.11519078473722102,
|
||
|
|
"grad_norm": 1.9296875,
|
||
|
|
"learning_rate": 2.8664921465968587e-05,
|
||
|
|
"loss": 1.2975,
|
||
|
|
"mean_token_accuracy": 0.6899457449093461,
|
||
|
|
"num_tokens": 3222578.0,
|
||
|
|
"step": 220
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2868557438254355,
|
||
|
|
"epoch": 0.1204267294980038,
|
||
|
|
"grad_norm": 2.0625,
|
||
|
|
"learning_rate": 2.99738219895288e-05,
|
||
|
|
"loss": 1.2843,
|
||
|
|
"mean_token_accuracy": 0.6913154577836395,
|
||
|
|
"num_tokens": 3367648.0,
|
||
|
|
"step": 230
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2696480546146631,
|
||
|
|
"epoch": 0.12566267425878658,
|
||
|
|
"grad_norm": 2.171875,
|
||
|
|
"learning_rate": 3.1282722513089006e-05,
|
||
|
|
"loss": 1.271,
|
||
|
|
"mean_token_accuracy": 0.6934641852974892,
|
||
|
|
"num_tokens": 3513611.0,
|
||
|
|
"step": 240
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2611329367384314,
|
||
|
|
"epoch": 0.13089861901956934,
|
||
|
|
"grad_norm": 2.515625,
|
||
|
|
"learning_rate": 3.2591623036649216e-05,
|
||
|
|
"loss": 1.27,
|
||
|
|
"mean_token_accuracy": 0.6962274318560958,
|
||
|
|
"num_tokens": 3654075.0,
|
||
|
|
"step": 250
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.3185498464852572,
|
||
|
|
"epoch": 0.13613456378035213,
|
||
|
|
"grad_norm": 2.0625,
|
||
|
|
"learning_rate": 3.3900523560209426e-05,
|
||
|
|
"loss": 1.3081,
|
||
|
|
"mean_token_accuracy": 0.6832702022045851,
|
||
|
|
"num_tokens": 3799547.0,
|
||
|
|
"step": 260
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2299459297209978,
|
||
|
|
"epoch": 0.1413705085411349,
|
||
|
|
"grad_norm": 2.09375,
|
||
|
|
"learning_rate": 3.5209424083769635e-05,
|
||
|
|
"loss": 1.2355,
|
||
|
|
"mean_token_accuracy": 0.6997983153909445,
|
||
|
|
"num_tokens": 3940094.0,
|
||
|
|
"step": 270
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2170744601637125,
|
||
|
|
"epoch": 0.14660645330191766,
|
||
|
|
"grad_norm": 1.9609375,
|
||
|
|
"learning_rate": 3.6518324607329845e-05,
|
||
|
|
"loss": 1.2097,
|
||
|
|
"mean_token_accuracy": 0.7026064315810799,
|
||
|
|
"num_tokens": 4084450.0,
|
||
|
|
"step": 280
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2503085616976022,
|
||
|
|
"epoch": 0.15184239806270045,
|
||
|
|
"grad_norm": 1.9375,
|
||
|
|
"learning_rate": 3.7827225130890054e-05,
|
||
|
|
"loss": 1.2366,
|
||
|
|
"mean_token_accuracy": 0.6935763908550143,
|
||
|
|
"num_tokens": 4235094.0,
|
||
|
|
"step": 290
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2002368062734603,
|
||
|
|
"epoch": 0.1570783428234832,
|
||
|
|
"grad_norm": 2.953125,
|
||
|
|
"learning_rate": 3.9136125654450264e-05,
|
||
|
|
"loss": 1.1993,
|
||
|
|
"mean_token_accuracy": 0.7031140483915805,
|
||
|
|
"num_tokens": 4378493.0,
|
||
|
|
"step": 300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2433323854580522,
|
||
|
|
"epoch": 0.162314287584266,
|
||
|
|
"grad_norm": 1.828125,
|
||
|
|
"learning_rate": 4.044502617801047e-05,
|
||
|
|
"loss": 1.2437,
|
||
|
|
"mean_token_accuracy": 0.6959625506773591,
|
||
|
|
"num_tokens": 4513351.0,
|
||
|
|
"step": 310
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2667765196412801,
|
||
|
|
"epoch": 0.16755023234504876,
|
||
|
|
"grad_norm": 2.078125,
|
||
|
|
"learning_rate": 4.175392670157068e-05,
|
||
|
|
"loss": 1.2609,
|
||
|
|
"mean_token_accuracy": 0.6903780495747924,
|
||
|
|
"num_tokens": 4654221.0,
|
||
|
|
"step": 320
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.257903415709734,
|
||
|
|
"epoch": 0.17278617710583152,
|
||
|
|
"grad_norm": 2.171875,
|
||
|
|
"learning_rate": 4.306282722513089e-05,
|
||
|
|
"loss": 1.2591,
|
||
|
|
"mean_token_accuracy": 0.6927119480445981,
|
||
|
|
"num_tokens": 4793001.0,
|
||
|
|
"step": 330
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2474724128842354,
|
||
|
|
"epoch": 0.1780221218666143,
|
||
|
|
"grad_norm": 1.8828125,
|
||
|
|
"learning_rate": 4.43717277486911e-05,
|
||
|
|
"loss": 1.258,
|
||
|
|
"mean_token_accuracy": 0.6958453560248017,
|
||
|
|
"num_tokens": 4938800.0,
|
||
|
|
"step": 340
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.271824512630701,
|
||
|
|
"epoch": 0.18325806662739708,
|
||
|
|
"grad_norm": 2.3125,
|
||
|
|
"learning_rate": 4.568062827225131e-05,
|
||
|
|
"loss": 1.2777,
|
||
|
|
"mean_token_accuracy": 0.6864122781902552,
|
||
|
|
"num_tokens": 5088676.0,
|
||
|
|
"step": 350
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.206870013102889,
|
||
|
|
"epoch": 0.18849401138817987,
|
||
|
|
"grad_norm": 2.1875,
|
||
|
|
"learning_rate": 4.698952879581152e-05,
|
||
|
|
"loss": 1.2145,
|
||
|
|
"mean_token_accuracy": 0.7025035681203008,
|
||
|
|
"num_tokens": 5233017.0,
|
||
|
|
"step": 360
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2822908700443805,
|
||
|
|
"epoch": 0.19372995614896263,
|
||
|
|
"grad_norm": 1.6953125,
|
||
|
|
"learning_rate": 4.829842931937173e-05,
|
||
|
|
"loss": 1.281,
|
||
|
|
"mean_token_accuracy": 0.685633241944015,
|
||
|
|
"num_tokens": 5383911.0,
|
||
|
|
"step": 370
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2813241746276618,
|
||
|
|
"epoch": 0.1989659009097454,
|
||
|
|
"grad_norm": 2.0,
|
||
|
|
"learning_rate": 4.960732984293194e-05,
|
||
|
|
"loss": 1.2896,
|
||
|
|
"mean_token_accuracy": 0.6846682282164693,
|
||
|
|
"num_tokens": 5532300.0,
|
||
|
|
"step": 380
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2817171201109887,
|
||
|
|
"epoch": 0.20420184567052818,
|
||
|
|
"grad_norm": 1.8359375,
|
||
|
|
"learning_rate": 4.999948856244768e-05,
|
||
|
|
"loss": 1.2811,
|
||
|
|
"mean_token_accuracy": 0.6898531707003712,
|
||
|
|
"num_tokens": 5673323.0,
|
||
|
|
"step": 390
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2512086292728781,
|
||
|
|
"epoch": 0.20943779043131094,
|
||
|
|
"grad_norm": 2.21875,
|
||
|
|
"learning_rate": 4.9996983612565773e-05,
|
||
|
|
"loss": 1.2605,
|
||
|
|
"mean_token_accuracy": 0.690356932580471,
|
||
|
|
"num_tokens": 5818798.0,
|
||
|
|
"step": 400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2130850929766894,
|
||
|
|
"epoch": 0.21467373519209373,
|
||
|
|
"grad_norm": 1.6640625,
|
||
|
|
"learning_rate": 4.999239142174581e-05,
|
||
|
|
"loss": 1.2191,
|
||
|
|
"mean_token_accuracy": 0.6989637283608318,
|
||
|
|
"num_tokens": 5967139.0,
|
||
|
|
"step": 410
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2118938906118273,
|
||
|
|
"epoch": 0.2199096799528765,
|
||
|
|
"grad_norm": 1.96875,
|
||
|
|
"learning_rate": 4.99857123734344e-05,
|
||
|
|
"loss": 1.2075,
|
||
|
|
"mean_token_accuracy": 0.6999017883092165,
|
||
|
|
"num_tokens": 6125485.0,
|
||
|
|
"step": 420
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2652900835499168,
|
||
|
|
"epoch": 0.22514562471365926,
|
||
|
|
"grad_norm": 1.9765625,
|
||
|
|
"learning_rate": 4.9976947025330155e-05,
|
||
|
|
"loss": 1.2729,
|
||
|
|
"mean_token_accuracy": 0.6881730291992426,
|
||
|
|
"num_tokens": 6271940.0,
|
||
|
|
"step": 430
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2763973344117403,
|
||
|
|
"epoch": 0.23038156947444205,
|
||
|
|
"grad_norm": 2.125,
|
||
|
|
"learning_rate": 4.9966096109337125e-05,
|
||
|
|
"loss": 1.3224,
|
||
|
|
"mean_token_accuracy": 0.6868822824209928,
|
||
|
|
"num_tokens": 6415021.0,
|
||
|
|
"step": 440
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2256551414728165,
|
||
|
|
"epoch": 0.2356175142352248,
|
||
|
|
"grad_norm": 1.6796875,
|
||
|
|
"learning_rate": 4.995316053150366e-05,
|
||
|
|
"loss": 1.2271,
|
||
|
|
"mean_token_accuracy": 0.6921151876449585,
|
||
|
|
"num_tokens": 6558653.0,
|
||
|
|
"step": 450
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.233339687436819,
|
||
|
|
"epoch": 0.2408534589960076,
|
||
|
|
"grad_norm": 2.34375,
|
||
|
|
"learning_rate": 4.993814137194681e-05,
|
||
|
|
"loss": 1.2679,
|
||
|
|
"mean_token_accuracy": 0.6938443537801504,
|
||
|
|
"num_tokens": 6698611.0,
|
||
|
|
"step": 460
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2777669046074152,
|
||
|
|
"epoch": 0.24608940375679036,
|
||
|
|
"grad_norm": 1.6640625,
|
||
|
|
"learning_rate": 4.9921039884762057e-05,
|
||
|
|
"loss": 1.2873,
|
||
|
|
"mean_token_accuracy": 0.6864250931888819,
|
||
|
|
"num_tokens": 6847385.0,
|
||
|
|
"step": 470
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.197480170428753,
|
||
|
|
"epoch": 0.25132534851757315,
|
||
|
|
"grad_norm": 2.03125,
|
||
|
|
"learning_rate": 4.9901857497918655e-05,
|
||
|
|
"loss": 1.2159,
|
||
|
|
"mean_token_accuracy": 0.7006905306130647,
|
||
|
|
"num_tokens": 6990503.0,
|
||
|
|
"step": 480
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2612487450242043,
|
||
|
|
"epoch": 0.2565612932783559,
|
||
|
|
"grad_norm": 2.140625,
|
||
|
|
"learning_rate": 4.98805958131404e-05,
|
||
|
|
"loss": 1.2667,
|
||
|
|
"mean_token_accuracy": 0.6904748784378171,
|
||
|
|
"num_tokens": 7138114.0,
|
||
|
|
"step": 490
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2380808498710394,
|
||
|
|
"epoch": 0.2617972380391387,
|
||
|
|
"grad_norm": 3.03125,
|
||
|
|
"learning_rate": 4.985725660577184e-05,
|
||
|
|
"loss": 1.2627,
|
||
|
|
"mean_token_accuracy": 0.693475303426385,
|
||
|
|
"num_tokens": 7295788.0,
|
||
|
|
"step": 500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2590212849900126,
|
||
|
|
"epoch": 0.26703318279992144,
|
||
|
|
"grad_norm": 2.03125,
|
||
|
|
"learning_rate": 4.983184182463009e-05,
|
||
|
|
"loss": 1.252,
|
||
|
|
"mean_token_accuracy": 0.6881766313686967,
|
||
|
|
"num_tokens": 7431663.0,
|
||
|
|
"step": 510
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2515497665852309,
|
||
|
|
"epoch": 0.27226912756070426,
|
||
|
|
"grad_norm": 3.703125,
|
||
|
|
"learning_rate": 4.980435359184204e-05,
|
||
|
|
"loss": 1.2831,
|
||
|
|
"mean_token_accuracy": 0.693330561555922,
|
||
|
|
"num_tokens": 7569801.0,
|
||
|
|
"step": 520
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.184147422760725,
|
||
|
|
"epoch": 0.277505072321487,
|
||
|
|
"grad_norm": 1.6328125,
|
||
|
|
"learning_rate": 4.977479420266723e-05,
|
||
|
|
"loss": 1.1646,
|
||
|
|
"mean_token_accuracy": 0.703077656775713,
|
||
|
|
"num_tokens": 7713092.0,
|
||
|
|
"step": 530
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2746197815984488,
|
||
|
|
"epoch": 0.2827410170822698,
|
||
|
|
"grad_norm": 2.0625,
|
||
|
|
"learning_rate": 4.974316612530615e-05,
|
||
|
|
"loss": 1.2765,
|
||
|
|
"mean_token_accuracy": 0.6863818326964974,
|
||
|
|
"num_tokens": 7854137.0,
|
||
|
|
"step": 540
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2566815540194511,
|
||
|
|
"epoch": 0.28797696184305255,
|
||
|
|
"grad_norm": 1.90625,
|
||
|
|
"learning_rate": 4.970947200069416e-05,
|
||
|
|
"loss": 1.2648,
|
||
|
|
"mean_token_accuracy": 0.6892008159309626,
|
||
|
|
"num_tokens": 7992676.0,
|
||
|
|
"step": 550
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2914229419082404,
|
||
|
|
"epoch": 0.2932129066038353,
|
||
|
|
"grad_norm": 1.640625,
|
||
|
|
"learning_rate": 4.967371464228096e-05,
|
||
|
|
"loss": 1.3179,
|
||
|
|
"mean_token_accuracy": 0.6854391321539879,
|
||
|
|
"num_tokens": 8149846.0,
|
||
|
|
"step": 560
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2183680593967439,
|
||
|
|
"epoch": 0.29844885136461813,
|
||
|
|
"grad_norm": 1.8046875,
|
||
|
|
"learning_rate": 4.963589703579569e-05,
|
||
|
|
"loss": 1.24,
|
||
|
|
"mean_token_accuracy": 0.6942471470683813,
|
||
|
|
"num_tokens": 8288016.0,
|
||
|
|
"step": 570
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.3149288706481457,
|
||
|
|
"epoch": 0.3036847961254009,
|
||
|
|
"grad_norm": 1.75,
|
||
|
|
"learning_rate": 4.959602233899762e-05,
|
||
|
|
"loss": 1.3198,
|
||
|
|
"mean_token_accuracy": 0.679352731257677,
|
||
|
|
"num_tokens": 8435936.0,
|
||
|
|
"step": 580
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.234706364199519,
|
||
|
|
"epoch": 0.30892074088618365,
|
||
|
|
"grad_norm": 1.7265625,
|
||
|
|
"learning_rate": 4.955409388141243e-05,
|
||
|
|
"loss": 1.2742,
|
||
|
|
"mean_token_accuracy": 0.6961829710751772,
|
||
|
|
"num_tokens": 8576679.0,
|
||
|
|
"step": 590
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2482819214463234,
|
||
|
|
"epoch": 0.3141566856469664,
|
||
|
|
"grad_norm": 2.1875,
|
||
|
|
"learning_rate": 4.9510115164054297e-05,
|
||
|
|
"loss": 1.2703,
|
||
|
|
"mean_token_accuracy": 0.6918804241344333,
|
||
|
|
"num_tokens": 8724159.0,
|
||
|
|
"step": 600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2805782459676265,
|
||
|
|
"epoch": 0.3193926304077492,
|
||
|
|
"grad_norm": 1.7578125,
|
||
|
|
"learning_rate": 4.946408985913344e-05,
|
||
|
|
"loss": 1.3043,
|
||
|
|
"mean_token_accuracy": 0.6854454703629017,
|
||
|
|
"num_tokens": 8871635.0,
|
||
|
|
"step": 610
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2219614367932081,
|
||
|
|
"epoch": 0.324628575168532,
|
||
|
|
"grad_norm": 1.578125,
|
||
|
|
"learning_rate": 4.941602180974958e-05,
|
||
|
|
"loss": 1.2703,
|
||
|
|
"mean_token_accuracy": 0.6931755296885967,
|
||
|
|
"num_tokens": 9025175.0,
|
||
|
|
"step": 620
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2477784302085637,
|
||
|
|
"epoch": 0.32986451992931476,
|
||
|
|
"grad_norm": 1.5625,
|
||
|
|
"learning_rate": 4.9365915029571007e-05,
|
||
|
|
"loss": 1.2585,
|
||
|
|
"mean_token_accuracy": 0.6917018702253699,
|
||
|
|
"num_tokens": 9162826.0,
|
||
|
|
"step": 630
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.230723000690341,
|
||
|
|
"epoch": 0.3351004646900975,
|
||
|
|
"grad_norm": 2.140625,
|
||
|
|
"learning_rate": 4.9313773702499455e-05,
|
||
|
|
"loss": 1.2391,
|
||
|
|
"mean_token_accuracy": 0.6946986148133873,
|
||
|
|
"num_tokens": 9312989.0,
|
||
|
|
"step": 640
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2285594891756773,
|
||
|
|
"epoch": 0.3403364094508803,
|
||
|
|
"grad_norm": 1.7578125,
|
||
|
|
"learning_rate": 4.925960218232073e-05,
|
||
|
|
"loss": 1.2367,
|
||
|
|
"mean_token_accuracy": 0.6967854388058186,
|
||
|
|
"num_tokens": 9455995.0,
|
||
|
|
"step": 650
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.3053442865610123,
|
||
|
|
"epoch": 0.34557235421166305,
|
||
|
|
"grad_norm": 2.328125,
|
||
|
|
"learning_rate": 4.920340499234116e-05,
|
||
|
|
"loss": 1.311,
|
||
|
|
"mean_token_accuracy": 0.6871177144348621,
|
||
|
|
"num_tokens": 9595268.0,
|
||
|
|
"step": 660
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2352213632315396,
|
||
|
|
"epoch": 0.35080829897244586,
|
||
|
|
"grad_norm": 2.125,
|
||
|
|
"learning_rate": 4.914518682500995e-05,
|
||
|
|
"loss": 1.2577,
|
||
|
|
"mean_token_accuracy": 0.69151939060539,
|
||
|
|
"num_tokens": 9737073.0,
|
||
|
|
"step": 670
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2220519341528415,
|
||
|
|
"epoch": 0.3560442437332286,
|
||
|
|
"grad_norm": 2.0625,
|
||
|
|
"learning_rate": 4.908495254152731e-05,
|
||
|
|
"loss": 1.2419,
|
||
|
|
"mean_token_accuracy": 0.7000335277989507,
|
||
|
|
"num_tokens": 9880500.0,
|
||
|
|
"step": 680
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2571235705167054,
|
||
|
|
"epoch": 0.3612801884940114,
|
||
|
|
"grad_norm": 1.890625,
|
||
|
|
"learning_rate": 4.902270717143859e-05,
|
||
|
|
"loss": 1.2772,
|
||
|
|
"mean_token_accuracy": 0.6950660437345505,
|
||
|
|
"num_tokens": 10029677.0,
|
||
|
|
"step": 690
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1837443890050054,
|
||
|
|
"epoch": 0.36651613325479415,
|
||
|
|
"grad_norm": 2.125,
|
||
|
|
"learning_rate": 4.895845591221426e-05,
|
||
|
|
"loss": 1.1883,
|
||
|
|
"mean_token_accuracy": 0.7034358236938715,
|
||
|
|
"num_tokens": 10169257.0,
|
||
|
|
"step": 700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1898296054452657,
|
||
|
|
"epoch": 0.3717520780155769,
|
||
|
|
"grad_norm": 1.96875,
|
||
|
|
"learning_rate": 4.8892204128816e-05,
|
||
|
|
"loss": 1.2018,
|
||
|
|
"mean_token_accuracy": 0.6991554461419582,
|
||
|
|
"num_tokens": 10312298.0,
|
||
|
|
"step": 710
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2473948691040277,
|
||
|
|
"epoch": 0.37698802277635973,
|
||
|
|
"grad_norm": 1.75,
|
||
|
|
"learning_rate": 4.882395735324864e-05,
|
||
|
|
"loss": 1.2845,
|
||
|
|
"mean_token_accuracy": 0.6941293969750404,
|
||
|
|
"num_tokens": 10455848.0,
|
||
|
|
"step": 720
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.217660278454423,
|
||
|
|
"epoch": 0.3822239675371425,
|
||
|
|
"grad_norm": 1.921875,
|
||
|
|
"learning_rate": 4.87537212840983e-05,
|
||
|
|
"loss": 1.2161,
|
||
|
|
"mean_token_accuracy": 0.6975388413295149,
|
||
|
|
"num_tokens": 10582901.0,
|
||
|
|
"step": 730
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2198585540056228,
|
||
|
|
"epoch": 0.38745991229792526,
|
||
|
|
"grad_norm": 1.8671875,
|
||
|
|
"learning_rate": 4.8681501786056544e-05,
|
||
|
|
"loss": 1.2394,
|
||
|
|
"mean_token_accuracy": 0.6949820145964622,
|
||
|
|
"num_tokens": 10724617.0,
|
||
|
|
"step": 740
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2234349481761455,
|
||
|
|
"epoch": 0.392695857058708,
|
||
|
|
"grad_norm": 1.7734375,
|
||
|
|
"learning_rate": 4.860730488943068e-05,
|
||
|
|
"loss": 1.2335,
|
||
|
|
"mean_token_accuracy": 0.6962696801871061,
|
||
|
|
"num_tokens": 10869452.0,
|
||
|
|
"step": 750
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2394332230091094,
|
||
|
|
"epoch": 0.3979318018194908,
|
||
|
|
"grad_norm": 2.296875,
|
||
|
|
"learning_rate": 4.8531136789640216e-05,
|
||
|
|
"loss": 1.2645,
|
||
|
|
"mean_token_accuracy": 0.695160668157041,
|
||
|
|
"num_tokens": 11017465.0,
|
||
|
|
"step": 760
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.219304683059454,
|
||
|
|
"epoch": 0.4031677465802736,
|
||
|
|
"grad_norm": 1.8125,
|
||
|
|
"learning_rate": 4.845300384669958e-05,
|
||
|
|
"loss": 1.2453,
|
||
|
|
"mean_token_accuracy": 0.6994039881974459,
|
||
|
|
"num_tokens": 11158876.0,
|
||
|
|
"step": 770
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.256004797667265,
|
||
|
|
"epoch": 0.40840369134105636,
|
||
|
|
"grad_norm": 1.953125,
|
||
|
|
"learning_rate": 4.837291258468701e-05,
|
||
|
|
"loss": 1.2915,
|
||
|
|
"mean_token_accuracy": 0.6878539452329278,
|
||
|
|
"num_tokens": 11297155.0,
|
||
|
|
"step": 780
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.201148072630167,
|
||
|
|
"epoch": 0.4136396361018391,
|
||
|
|
"grad_norm": 1.7734375,
|
||
|
|
"learning_rate": 4.8290869691199834e-05,
|
||
|
|
"loss": 1.22,
|
||
|
|
"mean_token_accuracy": 0.7018963057547808,
|
||
|
|
"num_tokens": 11439337.0,
|
||
|
|
"step": 790
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.233640456199646,
|
||
|
|
"epoch": 0.4188755808626219,
|
||
|
|
"grad_norm": 1.6796875,
|
||
|
|
"learning_rate": 4.820688201679605e-05,
|
||
|
|
"loss": 1.2485,
|
||
|
|
"mean_token_accuracy": 0.6938533913344145,
|
||
|
|
"num_tokens": 11585266.0,
|
||
|
|
"step": 800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2794130560010673,
|
||
|
|
"epoch": 0.42411152562340465,
|
||
|
|
"grad_norm": 1.75,
|
||
|
|
"learning_rate": 4.812095657442231e-05,
|
||
|
|
"loss": 1.2894,
|
||
|
|
"mean_token_accuracy": 0.6880532244220376,
|
||
|
|
"num_tokens": 11725643.0,
|
||
|
|
"step": 810
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2399780409410597,
|
||
|
|
"epoch": 0.42934747038418747,
|
||
|
|
"grad_norm": 1.7265625,
|
||
|
|
"learning_rate": 4.803310053882831e-05,
|
||
|
|
"loss": 1.2887,
|
||
|
|
"mean_token_accuracy": 0.6891330601647496,
|
||
|
|
"num_tokens": 11878982.0,
|
||
|
|
"step": 820
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2244506664574146,
|
||
|
|
"epoch": 0.43458341514497023,
|
||
|
|
"grad_norm": 1.6953125,
|
||
|
|
"learning_rate": 4.794332124596775e-05,
|
||
|
|
"loss": 1.2513,
|
||
|
|
"mean_token_accuracy": 0.6966191967949271,
|
||
|
|
"num_tokens": 12025844.0,
|
||
|
|
"step": 830
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2087435230612755,
|
||
|
|
"epoch": 0.439819359905753,
|
||
|
|
"grad_norm": 1.5,
|
||
|
|
"learning_rate": 4.7851626192385745e-05,
|
||
|
|
"loss": 1.2347,
|
||
|
|
"mean_token_accuracy": 0.7009090483188629,
|
||
|
|
"num_tokens": 12177345.0,
|
||
|
|
"step": 840
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2146162753924727,
|
||
|
|
"epoch": 0.44505530466653576,
|
||
|
|
"grad_norm": 1.8515625,
|
||
|
|
"learning_rate": 4.775802303459288e-05,
|
||
|
|
"loss": 1.2465,
|
||
|
|
"mean_token_accuracy": 0.6959697719663381,
|
||
|
|
"num_tokens": 12320998.0,
|
||
|
|
"step": 850
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.198814813606441,
|
||
|
|
"epoch": 0.4502912494273185,
|
||
|
|
"grad_norm": 1.875,
|
||
|
|
"learning_rate": 4.76625195884259e-05,
|
||
|
|
"loss": 1.2194,
|
||
|
|
"mean_token_accuracy": 0.6998335400596261,
|
||
|
|
"num_tokens": 12470540.0,
|
||
|
|
"step": 860
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1972925199195743,
|
||
|
|
"epoch": 0.45552719418810134,
|
||
|
|
"grad_norm": 1.5859375,
|
||
|
|
"learning_rate": 4.7565123828395064e-05,
|
||
|
|
"loss": 1.2199,
|
||
|
|
"mean_token_accuracy": 0.7027627993375063,
|
||
|
|
"num_tokens": 12605966.0,
|
||
|
|
"step": 870
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2037498267367481,
|
||
|
|
"epoch": 0.4607631389488841,
|
||
|
|
"grad_norm": 1.6484375,
|
||
|
|
"learning_rate": 4.7465843887018316e-05,
|
||
|
|
"loss": 1.2338,
|
||
|
|
"mean_token_accuracy": 0.7029284704476595,
|
||
|
|
"num_tokens": 12753176.0,
|
||
|
|
"step": 880
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2665604405105113,
|
||
|
|
"epoch": 0.46599908370966686,
|
||
|
|
"grad_norm": 1.953125,
|
||
|
|
"learning_rate": 4.736468805414218e-05,
|
||
|
|
"loss": 1.2826,
|
||
|
|
"mean_token_accuracy": 0.6867102902382612,
|
||
|
|
"num_tokens": 12891112.0,
|
||
|
|
"step": 890
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2413052493706345,
|
||
|
|
"epoch": 0.4712350284704496,
|
||
|
|
"grad_norm": 1.6953125,
|
||
|
|
"learning_rate": 4.72616647762496e-05,
|
||
|
|
"loss": 1.2708,
|
||
|
|
"mean_token_accuracy": 0.6939191322773695,
|
||
|
|
"num_tokens": 13037695.0,
|
||
|
|
"step": 900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2334762597456574,
|
||
|
|
"epoch": 0.4764709732312324,
|
||
|
|
"grad_norm": 1.8984375,
|
||
|
|
"learning_rate": 4.7156782655754625e-05,
|
||
|
|
"loss": 1.2801,
|
||
|
|
"mean_token_accuracy": 0.6962197717279196,
|
||
|
|
"num_tokens": 13188396.0,
|
||
|
|
"step": 910
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2926313485950232,
|
||
|
|
"epoch": 0.4817069179920152,
|
||
|
|
"grad_norm": 2.34375,
|
||
|
|
"learning_rate": 4.7050050450284147e-05,
|
||
|
|
"loss": 1.2973,
|
||
|
|
"mean_token_accuracy": 0.6834666855633259,
|
||
|
|
"num_tokens": 13334753.0,
|
||
|
|
"step": 920
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2605458820238709,
|
||
|
|
"epoch": 0.48694286275279797,
|
||
|
|
"grad_norm": 1.6953125,
|
||
|
|
"learning_rate": 4.6941477071946594e-05,
|
||
|
|
"loss": 1.273,
|
||
|
|
"mean_token_accuracy": 0.6905643936246634,
|
||
|
|
"num_tokens": 13481005.0,
|
||
|
|
"step": 930
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2309048125520348,
|
||
|
|
"epoch": 0.49217880751358073,
|
||
|
|
"grad_norm": 1.7421875,
|
||
|
|
"learning_rate": 4.683107158658781e-05,
|
||
|
|
"loss": 1.2438,
|
||
|
|
"mean_token_accuracy": 0.6907353041693568,
|
||
|
|
"num_tokens": 13618213.0,
|
||
|
|
"step": 940
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2219564571976662,
|
||
|
|
"epoch": 0.4974147522743635,
|
||
|
|
"grad_norm": 2.28125,
|
||
|
|
"learning_rate": 4.6718843213034067e-05,
|
||
|
|
"loss": 1.228,
|
||
|
|
"mean_token_accuracy": 0.6988438554108143,
|
||
|
|
"num_tokens": 13751965.0,
|
||
|
|
"step": 950
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1836062878370286,
|
||
|
|
"epoch": 0.5026506970351463,
|
||
|
|
"grad_norm": 6.375,
|
||
|
|
"learning_rate": 4.660480132232225e-05,
|
||
|
|
"loss": 1.2209,
|
||
|
|
"mean_token_accuracy": 0.6979983827099204,
|
||
|
|
"num_tokens": 13901640.0,
|
||
|
|
"step": 960
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1964640978723764,
|
||
|
|
"epoch": 0.507886641795929,
|
||
|
|
"grad_norm": 1.8984375,
|
||
|
|
"learning_rate": 4.648895543691741e-05,
|
||
|
|
"loss": 1.2005,
|
||
|
|
"mean_token_accuracy": 0.6987225420773029,
|
||
|
|
"num_tokens": 14047322.0,
|
||
|
|
"step": 970
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2625075351446866,
|
||
|
|
"epoch": 0.5131225865567118,
|
||
|
|
"grad_norm": 1.75,
|
||
|
|
"learning_rate": 4.637131522991764e-05,
|
||
|
|
"loss": 1.2557,
|
||
|
|
"mean_token_accuracy": 0.6897652598097921,
|
||
|
|
"num_tokens": 14188766.0,
|
||
|
|
"step": 980
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2458597056567668,
|
||
|
|
"epoch": 0.5183585313174947,
|
||
|
|
"grad_norm": 1.75,
|
||
|
|
"learning_rate": 4.625189052424638e-05,
|
||
|
|
"loss": 1.2676,
|
||
|
|
"mean_token_accuracy": 0.6945432106032967,
|
||
|
|
"num_tokens": 14329591.0,
|
||
|
|
"step": 990
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.245772442035377,
|
||
|
|
"epoch": 0.5235944760782774,
|
||
|
|
"grad_norm": 1.5625,
|
||
|
|
"learning_rate": 4.613069129183218e-05,
|
||
|
|
"loss": 1.2665,
|
||
|
|
"mean_token_accuracy": 0.6959560567513108,
|
||
|
|
"num_tokens": 14477258.0,
|
||
|
|
"step": 1000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2078047215938568,
|
||
|
|
"epoch": 0.5288304208390602,
|
||
|
|
"grad_norm": 1.6484375,
|
||
|
|
"learning_rate": 4.600772765277607e-05,
|
||
|
|
"loss": 1.2176,
|
||
|
|
"mean_token_accuracy": 0.697881168872118,
|
||
|
|
"num_tokens": 14633237.0,
|
||
|
|
"step": 1010
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2214165650308133,
|
||
|
|
"epoch": 0.5340663655998429,
|
||
|
|
"grad_norm": 1.765625,
|
||
|
|
"learning_rate": 4.588300987450652e-05,
|
||
|
|
"loss": 1.2345,
|
||
|
|
"mean_token_accuracy": 0.6999158889055253,
|
||
|
|
"num_tokens": 14776068.0,
|
||
|
|
"step": 1020
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2305742222815752,
|
||
|
|
"epoch": 0.5393023103606257,
|
||
|
|
"grad_norm": 2.34375,
|
||
|
|
"learning_rate": 4.575654837092214e-05,
|
||
|
|
"loss": 1.2398,
|
||
|
|
"mean_token_accuracy": 0.6949134254828095,
|
||
|
|
"num_tokens": 14913719.0,
|
||
|
|
"step": 1030
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2660383846610785,
|
||
|
|
"epoch": 0.5445382551214085,
|
||
|
|
"grad_norm": 1.6640625,
|
||
|
|
"learning_rate": 4.5628353701522055e-05,
|
||
|
|
"loss": 1.2685,
|
||
|
|
"mean_token_accuracy": 0.6927320031449199,
|
||
|
|
"num_tokens": 15066188.0,
|
||
|
|
"step": 1040
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2355303570628167,
|
||
|
|
"epoch": 0.5497741998821912,
|
||
|
|
"grad_norm": 1.8125,
|
||
|
|
"learning_rate": 4.5498436570524296e-05,
|
||
|
|
"loss": 1.2563,
|
||
|
|
"mean_token_accuracy": 0.6943851266056299,
|
||
|
|
"num_tokens": 15211662.0,
|
||
|
|
"step": 1050
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2411348339170218,
|
||
|
|
"epoch": 0.555010144642974,
|
||
|
|
"grad_norm": 1.6171875,
|
||
|
|
"learning_rate": 4.536680782597191e-05,
|
||
|
|
"loss": 1.2807,
|
||
|
|
"mean_token_accuracy": 0.6922256585210562,
|
||
|
|
"num_tokens": 15361477.0,
|
||
|
|
"step": 1060
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2601716944947838,
|
||
|
|
"epoch": 0.5602460894037568,
|
||
|
|
"grad_norm": 3.46875,
|
||
|
|
"learning_rate": 4.5233478458827176e-05,
|
||
|
|
"loss": 1.2964,
|
||
|
|
"mean_token_accuracy": 0.6930005580186844,
|
||
|
|
"num_tokens": 15512017.0,
|
||
|
|
"step": 1070
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.3123046960681677,
|
||
|
|
"epoch": 0.5654820341645396,
|
||
|
|
"grad_norm": 1.6796875,
|
||
|
|
"learning_rate": 4.509845960205389e-05,
|
||
|
|
"loss": 1.3435,
|
||
|
|
"mean_token_accuracy": 0.6821131203323603,
|
||
|
|
"num_tokens": 15666802.0,
|
||
|
|
"step": 1080
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1636217474937438,
|
||
|
|
"epoch": 0.5707179789253224,
|
||
|
|
"grad_norm": 1.9609375,
|
||
|
|
"learning_rate": 4.496176252968774e-05,
|
||
|
|
"loss": 1.1911,
|
||
|
|
"mean_token_accuracy": 0.7117271330207586,
|
||
|
|
"num_tokens": 15811306.0,
|
||
|
|
"step": 1090
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1874678194522859,
|
||
|
|
"epoch": 0.5759539236861051,
|
||
|
|
"grad_norm": 2.25,
|
||
|
|
"learning_rate": 4.4823398655894924e-05,
|
||
|
|
"loss": 1.1878,
|
||
|
|
"mean_token_accuracy": 0.7028247270733118,
|
||
|
|
"num_tokens": 15957078.0,
|
||
|
|
"step": 1100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.247788500599563,
|
||
|
|
"epoch": 0.5811898684468879,
|
||
|
|
"grad_norm": 2.0625,
|
||
|
|
"learning_rate": 4.468337953401908e-05,
|
||
|
|
"loss": 1.2483,
|
||
|
|
"mean_token_accuracy": 0.6906874619424344,
|
||
|
|
"num_tokens": 16099300.0,
|
||
|
|
"step": 1110
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2340633975341917,
|
||
|
|
"epoch": 0.5864258132076706,
|
||
|
|
"grad_norm": 1.734375,
|
||
|
|
"learning_rate": 4.45417168556166e-05,
|
||
|
|
"loss": 1.2385,
|
||
|
|
"mean_token_accuracy": 0.6969642581418156,
|
||
|
|
"num_tokens": 16241998.0,
|
||
|
|
"step": 1120
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1755719013512134,
|
||
|
|
"epoch": 0.5916617579684534,
|
||
|
|
"grad_norm": 1.7109375,
|
||
|
|
"learning_rate": 4.4398422449480356e-05,
|
||
|
|
"loss": 1.203,
|
||
|
|
"mean_token_accuracy": 0.7083542978391051,
|
||
|
|
"num_tokens": 16387522.0,
|
||
|
|
"step": 1130
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2200544375926257,
|
||
|
|
"epoch": 0.5968977027292363,
|
||
|
|
"grad_norm": 1.9375,
|
||
|
|
"learning_rate": 4.425350828065204e-05,
|
||
|
|
"loss": 1.2805,
|
||
|
|
"mean_token_accuracy": 0.6941867485642433,
|
||
|
|
"num_tokens": 16529425.0,
|
||
|
|
"step": 1140
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2577802315354347,
|
||
|
|
"epoch": 0.602133647490019,
|
||
|
|
"grad_norm": 2.265625,
|
||
|
|
"learning_rate": 4.410698644942303e-05,
|
||
|
|
"loss": 1.2505,
|
||
|
|
"mean_token_accuracy": 0.692890228703618,
|
||
|
|
"num_tokens": 16670311.0,
|
||
|
|
"step": 1150
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1618805171921849,
|
||
|
|
"epoch": 0.6073695922508018,
|
||
|
|
"grad_norm": 1.4921875,
|
||
|
|
"learning_rate": 4.395886919032406e-05,
|
||
|
|
"loss": 1.1903,
|
||
|
|
"mean_token_accuracy": 0.7042877223342657,
|
||
|
|
"num_tokens": 16813681.0,
|
||
|
|
"step": 1160
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2381837129592896,
|
||
|
|
"epoch": 0.6126055370115845,
|
||
|
|
"grad_norm": 1.9609375,
|
||
|
|
"learning_rate": 4.380916887110366e-05,
|
||
|
|
"loss": 1.2605,
|
||
|
|
"mean_token_accuracy": 0.6947620201855897,
|
||
|
|
"num_tokens": 16958167.0,
|
||
|
|
"step": 1170
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2540216479450463,
|
||
|
|
"epoch": 0.6178414817723673,
|
||
|
|
"grad_norm": 1.7890625,
|
||
|
|
"learning_rate": 4.365789799169539e-05,
|
||
|
|
"loss": 1.2586,
|
||
|
|
"mean_token_accuracy": 0.6949862573295832,
|
||
|
|
"num_tokens": 17101042.0,
|
||
|
|
"step": 1180
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2213780038058757,
|
||
|
|
"epoch": 0.6230774265331501,
|
||
|
|
"grad_norm": 1.6171875,
|
||
|
|
"learning_rate": 4.350506918317416e-05,
|
||
|
|
"loss": 1.2509,
|
||
|
|
"mean_token_accuracy": 0.6928936781361699,
|
||
|
|
"num_tokens": 17241395.0,
|
||
|
|
"step": 1190
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1887108445167542,
|
||
|
|
"epoch": 0.6283133712939328,
|
||
|
|
"grad_norm": 1.671875,
|
||
|
|
"learning_rate": 4.335069520670149e-05,
|
||
|
|
"loss": 1.2134,
|
||
|
|
"mean_token_accuracy": 0.7078650841489434,
|
||
|
|
"num_tokens": 17382621.0,
|
||
|
|
"step": 1200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1683177448809148,
|
||
|
|
"epoch": 0.6335493160547156,
|
||
|
|
"grad_norm": 1.8671875,
|
||
|
|
"learning_rate": 4.3194788952459996e-05,
|
||
|
|
"loss": 1.1862,
|
||
|
|
"mean_token_accuracy": 0.7075361222028732,
|
||
|
|
"num_tokens": 17524511.0,
|
||
|
|
"step": 1210
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2176016632467508,
|
||
|
|
"epoch": 0.6387852608154984,
|
||
|
|
"grad_norm": 1.5703125,
|
||
|
|
"learning_rate": 4.303736343857704e-05,
|
||
|
|
"loss": 1.2635,
|
||
|
|
"mean_token_accuracy": 0.6953268457204104,
|
||
|
|
"num_tokens": 17674442.0,
|
||
|
|
"step": 1220
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2287296935915948,
|
||
|
|
"epoch": 0.6440212055762812,
|
||
|
|
"grad_norm": 1.5234375,
|
||
|
|
"learning_rate": 4.2878431810037724e-05,
|
||
|
|
"loss": 1.239,
|
||
|
|
"mean_token_accuracy": 0.7013807725161314,
|
||
|
|
"num_tokens": 17827391.0,
|
||
|
|
"step": 1230
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.236463399976492,
|
||
|
|
"epoch": 0.649257150337064,
|
||
|
|
"grad_norm": 2.0,
|
||
|
|
"learning_rate": 4.27180073375873e-05,
|
||
|
|
"loss": 1.2571,
|
||
|
|
"mean_token_accuracy": 0.6894650906324387,
|
||
|
|
"num_tokens": 17972502.0,
|
||
|
|
"step": 1240
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1414937254041433,
|
||
|
|
"epoch": 0.6544930950978467,
|
||
|
|
"grad_norm": 1.640625,
|
||
|
|
"learning_rate": 4.255610341662304e-05,
|
||
|
|
"loss": 1.1546,
|
||
|
|
"mean_token_accuracy": 0.7119367253035307,
|
||
|
|
"num_tokens": 18115408.0,
|
||
|
|
"step": 1250
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.193948952294886,
|
||
|
|
"epoch": 0.6597290398586295,
|
||
|
|
"grad_norm": 1.8671875,
|
||
|
|
"learning_rate": 4.239273356607576e-05,
|
||
|
|
"loss": 1.2245,
|
||
|
|
"mean_token_accuracy": 0.7001152852550149,
|
||
|
|
"num_tokens": 18262489.0,
|
||
|
|
"step": 1260
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.246325920522213,
|
||
|
|
"epoch": 0.6649649846194122,
|
||
|
|
"grad_norm": 1.71875,
|
||
|
|
"learning_rate": 4.222791142728097e-05,
|
||
|
|
"loss": 1.2505,
|
||
|
|
"mean_token_accuracy": 0.6927321873605251,
|
||
|
|
"num_tokens": 18399068.0,
|
||
|
|
"step": 1270
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1776666756719352,
|
||
|
|
"epoch": 0.670200929380195,
|
||
|
|
"grad_norm": 1.8203125,
|
||
|
|
"learning_rate": 4.2061650762839825e-05,
|
||
|
|
"loss": 1.2147,
|
||
|
|
"mean_token_accuracy": 0.7039317097514868,
|
||
|
|
"num_tokens": 18543913.0,
|
||
|
|
"step": 1280
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2006115175783634,
|
||
|
|
"epoch": 0.6754368741409779,
|
||
|
|
"grad_norm": 1.5078125,
|
||
|
|
"learning_rate": 4.189396545546995e-05,
|
||
|
|
"loss": 1.2154,
|
||
|
|
"mean_token_accuracy": 0.7036881025880575,
|
||
|
|
"num_tokens": 18678858.0,
|
||
|
|
"step": 1290
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1709488430991768,
|
||
|
|
"epoch": 0.6806728189017606,
|
||
|
|
"grad_norm": 1.78125,
|
||
|
|
"learning_rate": 4.1724869506846267e-05,
|
||
|
|
"loss": 1.1949,
|
||
|
|
"mean_token_accuracy": 0.7089729970321059,
|
||
|
|
"num_tokens": 18834916.0,
|
||
|
|
"step": 1300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2211165010929108,
|
||
|
|
"epoch": 0.6859087636625434,
|
||
|
|
"grad_norm": 1.46875,
|
||
|
|
"learning_rate": 4.1554377036431816e-05,
|
||
|
|
"loss": 1.2437,
|
||
|
|
"mean_token_accuracy": 0.6957355309277773,
|
||
|
|
"num_tokens": 18987694.0,
|
||
|
|
"step": 1310
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2230720650404692,
|
||
|
|
"epoch": 0.6911447084233261,
|
||
|
|
"grad_norm": 1.7265625,
|
||
|
|
"learning_rate": 4.138250228029882e-05,
|
||
|
|
"loss": 1.2417,
|
||
|
|
"mean_token_accuracy": 0.6949638992547988,
|
||
|
|
"num_tokens": 19137646.0,
|
||
|
|
"step": 1320
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1824469189159572,
|
||
|
|
"epoch": 0.6963806531841089,
|
||
|
|
"grad_norm": 1.5546875,
|
||
|
|
"learning_rate": 4.120925958993994e-05,
|
||
|
|
"loss": 1.2038,
|
||
|
|
"mean_token_accuracy": 0.7080136310309172,
|
||
|
|
"num_tokens": 19285574.0,
|
||
|
|
"step": 1330
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1995828442275525,
|
||
|
|
"epoch": 0.7016165979448917,
|
||
|
|
"grad_norm": 1.8125,
|
||
|
|
"learning_rate": 4.103466343106999e-05,
|
||
|
|
"loss": 1.2104,
|
||
|
|
"mean_token_accuracy": 0.7012225743383169,
|
||
|
|
"num_tokens": 19432213.0,
|
||
|
|
"step": 1340
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2089691065251826,
|
||
|
|
"epoch": 0.7068525427056744,
|
||
|
|
"grad_norm": 1.8046875,
|
||
|
|
"learning_rate": 4.0858728382417966e-05,
|
||
|
|
"loss": 1.2252,
|
||
|
|
"mean_token_accuracy": 0.6981374306604267,
|
||
|
|
"num_tokens": 19586405.0,
|
||
|
|
"step": 1350
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2061294008046388,
|
||
|
|
"epoch": 0.7120884874664573,
|
||
|
|
"grad_norm": 1.6015625,
|
||
|
|
"learning_rate": 4.06814691345098e-05,
|
||
|
|
"loss": 1.2052,
|
||
|
|
"mean_token_accuracy": 0.6995515301823616,
|
||
|
|
"num_tokens": 19736389.0,
|
||
|
|
"step": 1360
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2196388389915227,
|
||
|
|
"epoch": 0.71732443222724,
|
||
|
|
"grad_norm": 1.6953125,
|
||
|
|
"learning_rate": 4.0502900488441706e-05,
|
||
|
|
"loss": 1.2295,
|
||
|
|
"mean_token_accuracy": 0.6974983751773834,
|
||
|
|
"num_tokens": 19884434.0,
|
||
|
|
"step": 1370
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1925444403663277,
|
||
|
|
"epoch": 0.7225603769880228,
|
||
|
|
"grad_norm": 1.65625,
|
||
|
|
"learning_rate": 4.032303735464422e-05,
|
||
|
|
"loss": 1.1885,
|
||
|
|
"mean_token_accuracy": 0.7031191129237413,
|
||
|
|
"num_tokens": 20019339.0,
|
||
|
|
"step": 1380
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.184141149930656,
|
||
|
|
"epoch": 0.7277963217488056,
|
||
|
|
"grad_norm": 1.515625,
|
||
|
|
"learning_rate": 4.0141894751637264e-05,
|
||
|
|
"loss": 1.2061,
|
||
|
|
"mean_token_accuracy": 0.7056893218308687,
|
||
|
|
"num_tokens": 20172036.0,
|
||
|
|
"step": 1390
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.18991824015975,
|
||
|
|
"epoch": 0.7330322665095883,
|
||
|
|
"grad_norm": 1.8125,
|
||
|
|
"learning_rate": 3.995948780477605e-05,
|
||
|
|
"loss": 1.1945,
|
||
|
|
"mean_token_accuracy": 0.7036307500675321,
|
||
|
|
"num_tokens": 20316279.0,
|
||
|
|
"step": 1400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.221291032806039,
|
||
|
|
"epoch": 0.7382682112703711,
|
||
|
|
"grad_norm": 2.234375,
|
||
|
|
"learning_rate": 3.977583174498816e-05,
|
||
|
|
"loss": 1.2229,
|
||
|
|
"mean_token_accuracy": 0.6996599985286593,
|
||
|
|
"num_tokens": 20457700.0,
|
||
|
|
"step": 1410
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2056787729263305,
|
||
|
|
"epoch": 0.7435041560311538,
|
||
|
|
"grad_norm": 1.9765625,
|
||
|
|
"learning_rate": 3.959094190750172e-05,
|
||
|
|
"loss": 1.2209,
|
||
|
|
"mean_token_accuracy": 0.699485157802701,
|
||
|
|
"num_tokens": 20601526.0,
|
||
|
|
"step": 1420
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.214169954136014,
|
||
|
|
"epoch": 0.7487401007919366,
|
||
|
|
"grad_norm": 1.640625,
|
||
|
|
"learning_rate": 3.940483373056498e-05,
|
||
|
|
"loss": 1.2251,
|
||
|
|
"mean_token_accuracy": 0.7045085027813911,
|
||
|
|
"num_tokens": 20740478.0,
|
||
|
|
"step": 1430
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2199758583679796,
|
||
|
|
"epoch": 0.7539760455527195,
|
||
|
|
"grad_norm": 1.421875,
|
||
|
|
"learning_rate": 3.921752275415712e-05,
|
||
|
|
"loss": 1.232,
|
||
|
|
"mean_token_accuracy": 0.6983612652868032,
|
||
|
|
"num_tokens": 20889352.0,
|
||
|
|
"step": 1440
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1763014759868384,
|
||
|
|
"epoch": 0.7592119903135022,
|
||
|
|
"grad_norm": 1.6875,
|
||
|
|
"learning_rate": 3.902902461869079e-05,
|
||
|
|
"loss": 1.1852,
|
||
|
|
"mean_token_accuracy": 0.706533107161522,
|
||
|
|
"num_tokens": 21047325.0,
|
||
|
|
"step": 1450
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2274865956045686,
|
||
|
|
"epoch": 0.764447935074285,
|
||
|
|
"grad_norm": 1.515625,
|
||
|
|
"learning_rate": 3.883935506370605e-05,
|
||
|
|
"loss": 1.2559,
|
||
|
|
"mean_token_accuracy": 0.6984432989731431,
|
||
|
|
"num_tokens": 21193313.0,
|
||
|
|
"step": 1460
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2031854771077632,
|
||
|
|
"epoch": 0.7696838798350677,
|
||
|
|
"grad_norm": 1.4140625,
|
||
|
|
"learning_rate": 3.864852992655617e-05,
|
||
|
|
"loss": 1.2106,
|
||
|
|
"mean_token_accuracy": 0.7018306776881218,
|
||
|
|
"num_tokens": 21340786.0,
|
||
|
|
"step": 1470
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2027717839926482,
|
||
|
|
"epoch": 0.7749198245958505,
|
||
|
|
"grad_norm": 1.671875,
|
||
|
|
"learning_rate": 3.845656514108515e-05,
|
||
|
|
"loss": 1.22,
|
||
|
|
"mean_token_accuracy": 0.7030729129910469,
|
||
|
|
"num_tokens": 21484126.0,
|
||
|
|
"step": 1480
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1977868607267737,
|
||
|
|
"epoch": 0.7801557693566333,
|
||
|
|
"grad_norm": 1.5,
|
||
|
|
"learning_rate": 3.8263476736297374e-05,
|
||
|
|
"loss": 1.1941,
|
||
|
|
"mean_token_accuracy": 0.7007357392460107,
|
||
|
|
"num_tokens": 21629712.0,
|
||
|
|
"step": 1490
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.280288253352046,
|
||
|
|
"epoch": 0.785391714117416,
|
||
|
|
"grad_norm": 1.53125,
|
||
|
|
"learning_rate": 3.806928083501906e-05,
|
||
|
|
"loss": 1.3073,
|
||
|
|
"mean_token_accuracy": 0.6855264658108353,
|
||
|
|
"num_tokens": 21769811.0,
|
||
|
|
"step": 1500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2234169896692038,
|
||
|
|
"epoch": 0.7906276588781989,
|
||
|
|
"grad_norm": 1.640625,
|
||
|
|
"learning_rate": 3.787399365255207e-05,
|
||
|
|
"loss": 1.2603,
|
||
|
|
"mean_token_accuracy": 0.6935047794133424,
|
||
|
|
"num_tokens": 21919583.0,
|
||
|
|
"step": 1510
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2192364005371927,
|
||
|
|
"epoch": 0.7958636036389816,
|
||
|
|
"grad_norm": 1.71875,
|
||
|
|
"learning_rate": 3.7677631495319956e-05,
|
||
|
|
"loss": 1.2092,
|
||
|
|
"mean_token_accuracy": 0.702324446476996,
|
||
|
|
"num_tokens": 22067375.0,
|
||
|
|
"step": 1520
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1928664781153202,
|
||
|
|
"epoch": 0.8010995483997644,
|
||
|
|
"grad_norm": 1.640625,
|
||
|
|
"learning_rate": 3.748021075950633e-05,
|
||
|
|
"loss": 1.2297,
|
||
|
|
"mean_token_accuracy": 0.6969239924103021,
|
||
|
|
"num_tokens": 22210682.0,
|
||
|
|
"step": 1530
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2294584538787603,
|
||
|
|
"epoch": 0.8063354931605472,
|
||
|
|
"grad_norm": 1.6328125,
|
||
|
|
"learning_rate": 3.728174792968582e-05,
|
||
|
|
"loss": 1.253,
|
||
|
|
"mean_token_accuracy": 0.6952543262392282,
|
||
|
|
"num_tokens": 22362042.0,
|
||
|
|
"step": 1540
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2315257797017694,
|
||
|
|
"epoch": 0.8115714379213299,
|
||
|
|
"grad_norm": 1.796875,
|
||
|
|
"learning_rate": 3.7082259577447605e-05,
|
||
|
|
"loss": 1.2633,
|
||
|
|
"mean_token_accuracy": 0.6920363411307335,
|
||
|
|
"num_tokens": 22518138.0,
|
||
|
|
"step": 1550
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2164895705878735,
|
||
|
|
"epoch": 0.8168073826821127,
|
||
|
|
"grad_norm": 1.4375,
|
||
|
|
"learning_rate": 3.688176236001168e-05,
|
||
|
|
"loss": 1.2215,
|
||
|
|
"mean_token_accuracy": 0.6953978851437569,
|
||
|
|
"num_tokens": 22670347.0,
|
||
|
|
"step": 1560
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1897184619680048,
|
||
|
|
"epoch": 0.8220433274428954,
|
||
|
|
"grad_norm": 1.6796875,
|
||
|
|
"learning_rate": 3.668027301883802e-05,
|
||
|
|
"loss": 1.2025,
|
||
|
|
"mean_token_accuracy": 0.7067706823348999,
|
||
|
|
"num_tokens": 22816471.0,
|
||
|
|
"step": 1570
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2356853460893036,
|
||
|
|
"epoch": 0.8272792722036783,
|
||
|
|
"grad_norm": 1.5078125,
|
||
|
|
"learning_rate": 3.6477808378228604e-05,
|
||
|
|
"loss": 1.2735,
|
||
|
|
"mean_token_accuracy": 0.6951639795675874,
|
||
|
|
"num_tokens": 22969371.0,
|
||
|
|
"step": 1580
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.172374564781785,
|
||
|
|
"epoch": 0.8325152169644611,
|
||
|
|
"grad_norm": 1.6640625,
|
||
|
|
"learning_rate": 3.6274385343922677e-05,
|
||
|
|
"loss": 1.1798,
|
||
|
|
"mean_token_accuracy": 0.7071554753929377,
|
||
|
|
"num_tokens": 23112802.0,
|
||
|
|
"step": 1590
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2456749143078922,
|
||
|
|
"epoch": 0.8377511617252438,
|
||
|
|
"grad_norm": 2.359375,
|
||
|
|
"learning_rate": 3.607002090168506e-05,
|
||
|
|
"loss": 1.2789,
|
||
|
|
"mean_token_accuracy": 0.6919172059744596,
|
||
|
|
"num_tokens": 23251431.0,
|
||
|
|
"step": 1600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2133430268615484,
|
||
|
|
"epoch": 0.8429871064860266,
|
||
|
|
"grad_norm": 1.75,
|
||
|
|
"learning_rate": 3.5864732115887866e-05,
|
||
|
|
"loss": 1.2482,
|
||
|
|
"mean_token_accuracy": 0.7028827562928199,
|
||
|
|
"num_tokens": 23398750.0,
|
||
|
|
"step": 1610
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2265673983842134,
|
||
|
|
"epoch": 0.8482230512468093,
|
||
|
|
"grad_norm": 1.5625,
|
||
|
|
"learning_rate": 3.565853612808562e-05,
|
||
|
|
"loss": 1.2666,
|
||
|
|
"mean_token_accuracy": 0.6953880734741688,
|
||
|
|
"num_tokens": 23541708.0,
|
||
|
|
"step": 1620
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1701237484812737,
|
||
|
|
"epoch": 0.8534589960075921,
|
||
|
|
"grad_norm": 1.609375,
|
||
|
|
"learning_rate": 3.545145015558399e-05,
|
||
|
|
"loss": 1.1376,
|
||
|
|
"mean_token_accuracy": 0.7052776984870434,
|
||
|
|
"num_tokens": 23689211.0,
|
||
|
|
"step": 1630
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.227908807620406,
|
||
|
|
"epoch": 0.8586949407683749,
|
||
|
|
"grad_norm": 1.578125,
|
||
|
|
"learning_rate": 3.524349149000206e-05,
|
||
|
|
"loss": 1.2574,
|
||
|
|
"mean_token_accuracy": 0.6968840681016445,
|
||
|
|
"num_tokens": 23830779.0,
|
||
|
|
"step": 1640
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1823246696963907,
|
||
|
|
"epoch": 0.8639308855291576,
|
||
|
|
"grad_norm": 1.7109375,
|
||
|
|
"learning_rate": 3.503467749582857e-05,
|
||
|
|
"loss": 1.176,
|
||
|
|
"mean_token_accuracy": 0.7053217653185129,
|
||
|
|
"num_tokens": 23966735.0,
|
||
|
|
"step": 1650
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1737437251955272,
|
||
|
|
"epoch": 0.8691668302899405,
|
||
|
|
"grad_norm": 1.6015625,
|
||
|
|
"learning_rate": 3.482502560897195e-05,
|
||
|
|
"loss": 1.202,
|
||
|
|
"mean_token_accuracy": 0.7063202302902937,
|
||
|
|
"num_tokens": 24114205.0,
|
||
|
|
"step": 1660
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2372199261561037,
|
||
|
|
"epoch": 0.8744027750507232,
|
||
|
|
"grad_norm": 1.46875,
|
||
|
|
"learning_rate": 3.4614553335304406e-05,
|
||
|
|
"loss": 1.2486,
|
||
|
|
"mean_token_accuracy": 0.694911016151309,
|
||
|
|
"num_tokens": 24257335.0,
|
||
|
|
"step": 1670
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2101770553737878,
|
||
|
|
"epoch": 0.879638719811506,
|
||
|
|
"grad_norm": 1.5,
|
||
|
|
"learning_rate": 3.440327824920022e-05,
|
||
|
|
"loss": 1.1971,
|
||
|
|
"mean_token_accuracy": 0.6996189601719379,
|
||
|
|
"num_tokens": 24405338.0,
|
||
|
|
"step": 1680
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1508656131103634,
|
||
|
|
"epoch": 0.8848746645722888,
|
||
|
|
"grad_norm": 1.6171875,
|
||
|
|
"learning_rate": 3.419121799206829e-05,
|
||
|
|
"loss": 1.17,
|
||
|
|
"mean_token_accuracy": 0.7079865211620927,
|
||
|
|
"num_tokens": 24550534.0,
|
||
|
|
"step": 1690
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2128825964406134,
|
||
|
|
"epoch": 0.8901106093330715,
|
||
|
|
"grad_norm": 1.71875,
|
||
|
|
"learning_rate": 3.3978390270879055e-05,
|
||
|
|
"loss": 1.248,
|
||
|
|
"mean_token_accuracy": 0.6985778672620654,
|
||
|
|
"num_tokens": 24699673.0,
|
||
|
|
"step": 1700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1805501360446216,
|
||
|
|
"epoch": 0.8953465540938543,
|
||
|
|
"grad_norm": 1.6171875,
|
||
|
|
"learning_rate": 3.3764812856686e-05,
|
||
|
|
"loss": 1.1703,
|
||
|
|
"mean_token_accuracy": 0.7011347938328981,
|
||
|
|
"num_tokens": 24848326.0,
|
||
|
|
"step": 1710
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.211562325246632,
|
||
|
|
"epoch": 0.900582498854637,
|
||
|
|
"grad_norm": 1.7421875,
|
||
|
|
"learning_rate": 3.355050358314172e-05,
|
||
|
|
"loss": 1.2425,
|
||
|
|
"mean_token_accuracy": 0.702686908468604,
|
||
|
|
"num_tokens": 24992122.0,
|
||
|
|
"step": 1720
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1974574619904161,
|
||
|
|
"epoch": 0.9058184436154199,
|
||
|
|
"grad_norm": 2.15625,
|
||
|
|
"learning_rate": 3.3335480345008905e-05,
|
||
|
|
"loss": 1.2259,
|
||
|
|
"mean_token_accuracy": 0.6980314027518034,
|
||
|
|
"num_tokens": 25139079.0,
|
||
|
|
"step": 1730
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2152755599468947,
|
||
|
|
"epoch": 0.9110543883762027,
|
||
|
|
"grad_norm": 1.8046875,
|
||
|
|
"learning_rate": 3.311976109666605e-05,
|
||
|
|
"loss": 1.251,
|
||
|
|
"mean_token_accuracy": 0.6992737432941795,
|
||
|
|
"num_tokens": 25286349.0,
|
||
|
|
"step": 1740
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2275765413418411,
|
||
|
|
"epoch": 0.9162903331369854,
|
||
|
|
"grad_norm": 1.4453125,
|
||
|
|
"learning_rate": 3.290336385060832e-05,
|
||
|
|
"loss": 1.2348,
|
||
|
|
"mean_token_accuracy": 0.6949001539498567,
|
||
|
|
"num_tokens": 25431706.0,
|
||
|
|
"step": 1750
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1663161270320415,
|
||
|
|
"epoch": 0.9215262778977682,
|
||
|
|
"grad_norm": 1.4765625,
|
||
|
|
"learning_rate": 3.268630667594348e-05,
|
||
|
|
"loss": 1.1854,
|
||
|
|
"mean_token_accuracy": 0.7103127352893353,
|
||
|
|
"num_tokens": 25575146.0,
|
||
|
|
"step": 1760
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2058376437053084,
|
||
|
|
"epoch": 0.9267622226585509,
|
||
|
|
"grad_norm": 1.4765625,
|
||
|
|
"learning_rate": 3.2468607696883146e-05,
|
||
|
|
"loss": 1.2425,
|
||
|
|
"mean_token_accuracy": 0.7038685705512762,
|
||
|
|
"num_tokens": 25726777.0,
|
||
|
|
"step": 1770
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.189345240779221,
|
||
|
|
"epoch": 0.9319981674193337,
|
||
|
|
"grad_norm": 2.03125,
|
||
|
|
"learning_rate": 3.225028509122944e-05,
|
||
|
|
"loss": 1.2287,
|
||
|
|
"mean_token_accuracy": 0.7018254602327942,
|
||
|
|
"num_tokens": 25875049.0,
|
||
|
|
"step": 1780
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1771747019141912,
|
||
|
|
"epoch": 0.9372341121801165,
|
||
|
|
"grad_norm": 1.515625,
|
||
|
|
"learning_rate": 3.2031357088857085e-05,
|
||
|
|
"loss": 1.1915,
|
||
|
|
"mean_token_accuracy": 0.7071957625448704,
|
||
|
|
"num_tokens": 26026216.0,
|
||
|
|
"step": 1790
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.220152474567294,
|
||
|
|
"epoch": 0.9424700569408992,
|
||
|
|
"grad_norm": 1.8828125,
|
||
|
|
"learning_rate": 3.181184197019127e-05,
|
||
|
|
"loss": 1.2299,
|
||
|
|
"mean_token_accuracy": 0.7012155883014202,
|
||
|
|
"num_tokens": 26170779.0,
|
||
|
|
"step": 1800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1854938926175236,
|
||
|
|
"epoch": 0.9477060017016821,
|
||
|
|
"grad_norm": 1.625,
|
||
|
|
"learning_rate": 3.159175806468126e-05,
|
||
|
|
"loss": 1.1874,
|
||
|
|
"mean_token_accuracy": 0.705037958547473,
|
||
|
|
"num_tokens": 26319322.0,
|
||
|
|
"step": 1810
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2158665416762233,
|
||
|
|
"epoch": 0.9529419464624648,
|
||
|
|
"grad_norm": 1.3984375,
|
||
|
|
"learning_rate": 3.1371123749269805e-05,
|
||
|
|
"loss": 1.2222,
|
||
|
|
"mean_token_accuracy": 0.6985156688839197,
|
||
|
|
"num_tokens": 26465154.0,
|
||
|
|
"step": 1820
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2151709901168943,
|
||
|
|
"epoch": 0.9581778912232476,
|
||
|
|
"grad_norm": 1.6953125,
|
||
|
|
"learning_rate": 3.114995744685877e-05,
|
||
|
|
"loss": 1.2667,
|
||
|
|
"mean_token_accuracy": 0.6974882191047073,
|
||
|
|
"num_tokens": 26611774.0,
|
||
|
|
"step": 1830
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1511426636949182,
|
||
|
|
"epoch": 0.9634138359840304,
|
||
|
|
"grad_norm": 1.765625,
|
||
|
|
"learning_rate": 3.092827762477074e-05,
|
||
|
|
"loss": 1.197,
|
||
|
|
"mean_token_accuracy": 0.7165371583774686,
|
||
|
|
"num_tokens": 26760701.0,
|
||
|
|
"step": 1840
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2074316812679171,
|
||
|
|
"epoch": 0.9686497807448131,
|
||
|
|
"grad_norm": 1.5234375,
|
||
|
|
"learning_rate": 3.070610279320707e-05,
|
||
|
|
"loss": 1.2485,
|
||
|
|
"mean_token_accuracy": 0.7016795247793197,
|
||
|
|
"num_tokens": 26902851.0,
|
||
|
|
"step": 1850
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1557184986770153,
|
||
|
|
"epoch": 0.9738857255055959,
|
||
|
|
"grad_norm": 1.5,
|
||
|
|
"learning_rate": 3.0483451503702264e-05,
|
||
|
|
"loss": 1.1973,
|
||
|
|
"mean_token_accuracy": 0.7114167800173163,
|
||
|
|
"num_tokens": 27053159.0,
|
||
|
|
"step": 1860
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.145219304971397,
|
||
|
|
"epoch": 0.9791216702663786,
|
||
|
|
"grad_norm": 1.6015625,
|
||
|
|
"learning_rate": 3.0260342347574915e-05,
|
||
|
|
"loss": 1.1371,
|
||
|
|
"mean_token_accuracy": 0.7116047518327833,
|
||
|
|
"num_tokens": 27195594.0,
|
||
|
|
"step": 1870
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1833709230646492,
|
||
|
|
"epoch": 0.9843576150271615,
|
||
|
|
"grad_norm": 1.578125,
|
||
|
|
"learning_rate": 3.003679395437536e-05,
|
||
|
|
"loss": 1.1913,
|
||
|
|
"mean_token_accuracy": 0.7063193745911122,
|
||
|
|
"num_tokens": 27335855.0,
|
||
|
|
"step": 1880
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2038549520075321,
|
||
|
|
"epoch": 0.9895935597879443,
|
||
|
|
"grad_norm": 1.7578125,
|
||
|
|
"learning_rate": 2.981282499033009e-05,
|
||
|
|
"loss": 1.2036,
|
||
|
|
"mean_token_accuracy": 0.6992362190037966,
|
||
|
|
"num_tokens": 27485225.0,
|
||
|
|
"step": 1890
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1822845570743084,
|
||
|
|
"epoch": 0.994829504548727,
|
||
|
|
"grad_norm": 1.6328125,
|
||
|
|
"learning_rate": 2.9588454156783163e-05,
|
||
|
|
"loss": 1.2295,
|
||
|
|
"mean_token_accuracy": 0.7060987044125795,
|
||
|
|
"num_tokens": 27634788.0,
|
||
|
|
"step": 1900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1678016036748886,
|
||
|
|
"epoch": 1.0,
|
||
|
|
"grad_norm": 1.96875,
|
||
|
|
"learning_rate": 2.9363700188634598e-05,
|
||
|
|
"loss": 1.1579,
|
||
|
|
"mean_token_accuracy": 0.7110490016167677,
|
||
|
|
"num_tokens": 27776648.0,
|
||
|
|
"step": 1910
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.0,
|
||
|
|
"eval_entropy": 1.1545482213497162,
|
||
|
|
"eval_loss": 1.1723113059997559,
|
||
|
|
"eval_mean_token_accuracy": 0.7067238150835037,
|
||
|
|
"eval_num_tokens": 27776648.0,
|
||
|
|
"eval_runtime": 60.9052,
|
||
|
|
"eval_samples_per_second": 32.838,
|
||
|
|
"eval_steps_per_second": 16.419,
|
||
|
|
"step": 1910
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 10,
|
||
|
|
"max_steps": 3820,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 2,
|
||
|
|
"save_steps": 500,
|
||
|
|
"stateful_callbacks": {
|
||
|
|
"TrainerControl": {
|
||
|
|
"args": {
|
||
|
|
"should_epoch_stop": false,
|
||
|
|
"should_evaluate": false,
|
||
|
|
"should_log": false,
|
||
|
|
"should_save": true,
|
||
|
|
"should_training_stop": false
|
||
|
|
},
|
||
|
|
"attributes": {}
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"total_flos": 2.962575601921106e+17,
|
||
|
|
"train_batch_size": 2,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|