345 lines
9.5 KiB
JSON
345 lines
9.5 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 1.0,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 313,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"entropy": 1.4538299292325974,
|
||
|
|
"epoch": 0.03201280512204882,
|
||
|
|
"grad_norm": 22.75,
|
||
|
|
"learning_rate": 5.806451612903226e-06,
|
||
|
|
"loss": 3.070576477050781,
|
||
|
|
"mean_token_accuracy": 0.4463906198740005,
|
||
|
|
"num_tokens": 39534.0,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.7499498546123504,
|
||
|
|
"epoch": 0.06402561024409764,
|
||
|
|
"grad_norm": 10.5,
|
||
|
|
"learning_rate": 1.2258064516129034e-05,
|
||
|
|
"loss": 2.6519420623779295,
|
||
|
|
"mean_token_accuracy": 0.46584000922739505,
|
||
|
|
"num_tokens": 79078.0,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 2.058299073576927,
|
||
|
|
"epoch": 0.09603841536614646,
|
||
|
|
"grad_norm": 6.53125,
|
||
|
|
"learning_rate": 1.870967741935484e-05,
|
||
|
|
"loss": 2.1608741760253904,
|
||
|
|
"mean_token_accuracy": 0.5163775213062763,
|
||
|
|
"num_tokens": 118225.0,
|
||
|
|
"step": 30
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.9837611675262452,
|
||
|
|
"epoch": 0.12805122048819528,
|
||
|
|
"grad_norm": 4.3125,
|
||
|
|
"learning_rate": 1.9432624113475178e-05,
|
||
|
|
"loss": 2.000233268737793,
|
||
|
|
"mean_token_accuracy": 0.5434215664863586,
|
||
|
|
"num_tokens": 158198.0,
|
||
|
|
"step": 40
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.8247230246663093,
|
||
|
|
"epoch": 0.1600640256102441,
|
||
|
|
"grad_norm": 3.9375,
|
||
|
|
"learning_rate": 1.872340425531915e-05,
|
||
|
|
"loss": 1.885169219970703,
|
||
|
|
"mean_token_accuracy": 0.5518886391073465,
|
||
|
|
"num_tokens": 197695.0,
|
||
|
|
"step": 50
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.820909820497036,
|
||
|
|
"epoch": 0.19207683073229292,
|
||
|
|
"grad_norm": 3.46875,
|
||
|
|
"learning_rate": 1.801418439716312e-05,
|
||
|
|
"loss": 1.8881864547729492,
|
||
|
|
"mean_token_accuracy": 0.5554269794374704,
|
||
|
|
"num_tokens": 238556.0,
|
||
|
|
"step": 60
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.8308037787675857,
|
||
|
|
"epoch": 0.22408963585434175,
|
||
|
|
"grad_norm": 3.46875,
|
||
|
|
"learning_rate": 1.7304964539007094e-05,
|
||
|
|
"loss": 1.8693714141845703,
|
||
|
|
"mean_token_accuracy": 0.5529689989984036,
|
||
|
|
"num_tokens": 278868.0,
|
||
|
|
"step": 70
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.7878538712859153,
|
||
|
|
"epoch": 0.25610244097639057,
|
||
|
|
"grad_norm": 3.6875,
|
||
|
|
"learning_rate": 1.6595744680851064e-05,
|
||
|
|
"loss": 1.847772216796875,
|
||
|
|
"mean_token_accuracy": 0.5583747569471598,
|
||
|
|
"num_tokens": 319510.0,
|
||
|
|
"step": 80
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.806380895525217,
|
||
|
|
"epoch": 0.28811524609843936,
|
||
|
|
"grad_norm": 3.578125,
|
||
|
|
"learning_rate": 1.5886524822695038e-05,
|
||
|
|
"loss": 1.8553556442260741,
|
||
|
|
"mean_token_accuracy": 0.562133489176631,
|
||
|
|
"num_tokens": 360080.0,
|
||
|
|
"step": 90
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.7894016727805138,
|
||
|
|
"epoch": 0.3201280512204882,
|
||
|
|
"grad_norm": 3.46875,
|
||
|
|
"learning_rate": 1.5177304964539008e-05,
|
||
|
|
"loss": 1.8526411056518555,
|
||
|
|
"mean_token_accuracy": 0.5623987793922425,
|
||
|
|
"num_tokens": 400106.0,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.863182120025158,
|
||
|
|
"epoch": 0.352140856342537,
|
||
|
|
"grad_norm": 3.34375,
|
||
|
|
"learning_rate": 1.4468085106382981e-05,
|
||
|
|
"loss": 1.8998908996582031,
|
||
|
|
"mean_token_accuracy": 0.5487093910574913,
|
||
|
|
"num_tokens": 442713.0,
|
||
|
|
"step": 110
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.8063384279608727,
|
||
|
|
"epoch": 0.38415366146458585,
|
||
|
|
"grad_norm": 3.453125,
|
||
|
|
"learning_rate": 1.3758865248226951e-05,
|
||
|
|
"loss": 1.8471385955810546,
|
||
|
|
"mean_token_accuracy": 0.557015884667635,
|
||
|
|
"num_tokens": 483468.0,
|
||
|
|
"step": 120
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.8770785540342332,
|
||
|
|
"epoch": 0.41616646658663464,
|
||
|
|
"grad_norm": 3.609375,
|
||
|
|
"learning_rate": 1.3049645390070925e-05,
|
||
|
|
"loss": 1.9237331390380858,
|
||
|
|
"mean_token_accuracy": 0.5443044405430555,
|
||
|
|
"num_tokens": 526232.0,
|
||
|
|
"step": 130
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.8102150335907936,
|
||
|
|
"epoch": 0.4481792717086835,
|
||
|
|
"grad_norm": 3.6875,
|
||
|
|
"learning_rate": 1.2340425531914895e-05,
|
||
|
|
"loss": 1.8544883728027344,
|
||
|
|
"mean_token_accuracy": 0.558148629963398,
|
||
|
|
"num_tokens": 568427.0,
|
||
|
|
"step": 140
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.7433957248926162,
|
||
|
|
"epoch": 0.4801920768307323,
|
||
|
|
"grad_norm": 3.28125,
|
||
|
|
"learning_rate": 1.1631205673758865e-05,
|
||
|
|
"loss": 1.7788505554199219,
|
||
|
|
"mean_token_accuracy": 0.5669799156486988,
|
||
|
|
"num_tokens": 608610.0,
|
||
|
|
"step": 150
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.788873042166233,
|
||
|
|
"epoch": 0.5122048819527811,
|
||
|
|
"grad_norm": 3.484375,
|
||
|
|
"learning_rate": 1.0921985815602838e-05,
|
||
|
|
"loss": 1.8689699172973633,
|
||
|
|
"mean_token_accuracy": 0.5592933889478445,
|
||
|
|
"num_tokens": 649517.0,
|
||
|
|
"step": 160
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.784371581673622,
|
||
|
|
"epoch": 0.54421768707483,
|
||
|
|
"grad_norm": 3.46875,
|
||
|
|
"learning_rate": 1.0212765957446808e-05,
|
||
|
|
"loss": 1.8266977310180663,
|
||
|
|
"mean_token_accuracy": 0.5599350105971098,
|
||
|
|
"num_tokens": 690274.0,
|
||
|
|
"step": 170
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.770761302113533,
|
||
|
|
"epoch": 0.5762304921968787,
|
||
|
|
"grad_norm": 3.359375,
|
||
|
|
"learning_rate": 9.503546099290782e-06,
|
||
|
|
"loss": 1.8199588775634765,
|
||
|
|
"mean_token_accuracy": 0.5642654895782471,
|
||
|
|
"num_tokens": 732104.0,
|
||
|
|
"step": 180
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.785500568151474,
|
||
|
|
"epoch": 0.6082432973189276,
|
||
|
|
"grad_norm": 3.359375,
|
||
|
|
"learning_rate": 8.794326241134753e-06,
|
||
|
|
"loss": 1.8203857421875,
|
||
|
|
"mean_token_accuracy": 0.5628614298999309,
|
||
|
|
"num_tokens": 772388.0,
|
||
|
|
"step": 190
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.7996786698698997,
|
||
|
|
"epoch": 0.6402561024409764,
|
||
|
|
"grad_norm": 3.234375,
|
||
|
|
"learning_rate": 8.085106382978723e-06,
|
||
|
|
"loss": 1.838487434387207,
|
||
|
|
"mean_token_accuracy": 0.5601454164832831,
|
||
|
|
"num_tokens": 814417.0,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.7707496628165245,
|
||
|
|
"epoch": 0.6722689075630253,
|
||
|
|
"grad_norm": 3.5625,
|
||
|
|
"learning_rate": 7.375886524822695e-06,
|
||
|
|
"loss": 1.8333843231201172,
|
||
|
|
"mean_token_accuracy": 0.5659072320908308,
|
||
|
|
"num_tokens": 855220.0,
|
||
|
|
"step": 210
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.849867083132267,
|
||
|
|
"epoch": 0.704281712685074,
|
||
|
|
"grad_norm": 3.25,
|
||
|
|
"learning_rate": 6.666666666666667e-06,
|
||
|
|
"loss": 1.9260578155517578,
|
||
|
|
"mean_token_accuracy": 0.5531776797026395,
|
||
|
|
"num_tokens": 898972.0,
|
||
|
|
"step": 220
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.7605908274650575,
|
||
|
|
"epoch": 0.7362945178071229,
|
||
|
|
"grad_norm": 3.4375,
|
||
|
|
"learning_rate": 5.957446808510638e-06,
|
||
|
|
"loss": 1.7972305297851563,
|
||
|
|
"mean_token_accuracy": 0.5678223472088575,
|
||
|
|
"num_tokens": 938284.0,
|
||
|
|
"step": 230
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.7362026132643222,
|
||
|
|
"epoch": 0.7683073229291717,
|
||
|
|
"grad_norm": 3.484375,
|
||
|
|
"learning_rate": 5.24822695035461e-06,
|
||
|
|
"loss": 1.7981218338012694,
|
||
|
|
"mean_token_accuracy": 0.5710738152265549,
|
||
|
|
"num_tokens": 977880.0,
|
||
|
|
"step": 240
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.79102204144001,
|
||
|
|
"epoch": 0.8003201280512204,
|
||
|
|
"grad_norm": 3.421875,
|
||
|
|
"learning_rate": 4.539007092198582e-06,
|
||
|
|
"loss": 1.8328744888305664,
|
||
|
|
"mean_token_accuracy": 0.5625536382198334,
|
||
|
|
"num_tokens": 1019549.0,
|
||
|
|
"step": 250
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.788681410253048,
|
||
|
|
"epoch": 0.8323329331732693,
|
||
|
|
"grad_norm": 3.375,
|
||
|
|
"learning_rate": 3.8297872340425535e-06,
|
||
|
|
"loss": 1.8353548049926758,
|
||
|
|
"mean_token_accuracy": 0.5642140224575997,
|
||
|
|
"num_tokens": 1060145.0,
|
||
|
|
"step": 260
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.8368026711046697,
|
||
|
|
"epoch": 0.8643457382953181,
|
||
|
|
"grad_norm": 3.28125,
|
||
|
|
"learning_rate": 3.120567375886525e-06,
|
||
|
|
"loss": 1.9220146179199218,
|
||
|
|
"mean_token_accuracy": 0.5521392654627562,
|
||
|
|
"num_tokens": 1102818.0,
|
||
|
|
"step": 270
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.7728961274027824,
|
||
|
|
"epoch": 0.896358543417367,
|
||
|
|
"grad_norm": 3.546875,
|
||
|
|
"learning_rate": 2.4113475177304965e-06,
|
||
|
|
"loss": 1.8067329406738282,
|
||
|
|
"mean_token_accuracy": 0.5586034987121821,
|
||
|
|
"num_tokens": 1143300.0,
|
||
|
|
"step": 280
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.7760732486844062,
|
||
|
|
"epoch": 0.9283713485394157,
|
||
|
|
"grad_norm": 3.328125,
|
||
|
|
"learning_rate": 1.7021276595744682e-06,
|
||
|
|
"loss": 1.8068740844726563,
|
||
|
|
"mean_token_accuracy": 0.5682683397084475,
|
||
|
|
"num_tokens": 1183341.0,
|
||
|
|
"step": 290
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.7398153685033322,
|
||
|
|
"epoch": 0.9603841536614646,
|
||
|
|
"grad_norm": 3.75,
|
||
|
|
"learning_rate": 9.929078014184399e-07,
|
||
|
|
"loss": 1.7799579620361328,
|
||
|
|
"mean_token_accuracy": 0.5732921231538057,
|
||
|
|
"num_tokens": 1223298.0,
|
||
|
|
"step": 300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.8165026500821113,
|
||
|
|
"epoch": 0.9923969587835134,
|
||
|
|
"grad_norm": 3.265625,
|
||
|
|
"learning_rate": 2.8368794326241136e-07,
|
||
|
|
"loss": 1.875440788269043,
|
||
|
|
"mean_token_accuracy": 0.5596334230154753,
|
||
|
|
"num_tokens": 1265400.0,
|
||
|
|
"step": 310
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 10,
|
||
|
|
"max_steps": 313,
|
||
|
|
"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": 1.2815731740622848e+16,
|
||
|
|
"train_batch_size": 2,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|