prototype_joint_trained / trainer_state.json
Serega6678's picture
Model save
acd72fd verified
raw
history blame
22.8 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"eval_steps": 500,
"global_step": 907,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 2.197802197802198e-07,
"loss": 0.998,
"step": 1
},
{
"epoch": 0.01,
"learning_rate": 1.098901098901099e-06,
"loss": 1.0162,
"step": 5
},
{
"epoch": 0.01,
"learning_rate": 2.197802197802198e-06,
"loss": 0.9881,
"step": 10
},
{
"epoch": 0.02,
"learning_rate": 3.2967032967032968e-06,
"loss": 1.0192,
"step": 15
},
{
"epoch": 0.02,
"learning_rate": 4.395604395604396e-06,
"loss": 1.0097,
"step": 20
},
{
"epoch": 0.03,
"learning_rate": 5.494505494505495e-06,
"loss": 1.0106,
"step": 25
},
{
"epoch": 0.03,
"learning_rate": 6.5934065934065935e-06,
"loss": 1.0132,
"step": 30
},
{
"epoch": 0.04,
"learning_rate": 7.692307692307694e-06,
"loss": 0.9951,
"step": 35
},
{
"epoch": 0.04,
"learning_rate": 8.791208791208792e-06,
"loss": 1.0027,
"step": 40
},
{
"epoch": 0.05,
"learning_rate": 9.890109890109892e-06,
"loss": 1.0178,
"step": 45
},
{
"epoch": 0.06,
"learning_rate": 1.098901098901099e-05,
"loss": 1.0272,
"step": 50
},
{
"epoch": 0.06,
"learning_rate": 1.2087912087912089e-05,
"loss": 1.0013,
"step": 55
},
{
"epoch": 0.07,
"learning_rate": 1.3186813186813187e-05,
"loss": 1.0244,
"step": 60
},
{
"epoch": 0.07,
"learning_rate": 1.4285714285714287e-05,
"loss": 1.0224,
"step": 65
},
{
"epoch": 0.08,
"learning_rate": 1.5384615384615387e-05,
"loss": 1.0138,
"step": 70
},
{
"epoch": 0.08,
"learning_rate": 1.6483516483516486e-05,
"loss": 1.0,
"step": 75
},
{
"epoch": 0.09,
"learning_rate": 1.7582417582417584e-05,
"loss": 1.0053,
"step": 80
},
{
"epoch": 0.09,
"learning_rate": 1.8681318681318682e-05,
"loss": 0.9828,
"step": 85
},
{
"epoch": 0.1,
"learning_rate": 1.9780219780219784e-05,
"loss": 1.0125,
"step": 90
},
{
"epoch": 0.1,
"learning_rate": 1.999881422898077e-05,
"loss": 0.9976,
"step": 95
},
{
"epoch": 0.11,
"learning_rate": 1.999399751616869e-05,
"loss": 1.0147,
"step": 100
},
{
"epoch": 0.12,
"learning_rate": 1.9985477534331582e-05,
"loss": 0.9959,
"step": 105
},
{
"epoch": 0.12,
"learning_rate": 1.9973257440542974e-05,
"loss": 0.9702,
"step": 110
},
{
"epoch": 0.13,
"learning_rate": 1.9957341762950346e-05,
"loss": 0.996,
"step": 115
},
{
"epoch": 0.13,
"learning_rate": 1.9937736399097282e-05,
"loss": 1.0323,
"step": 120
},
{
"epoch": 0.14,
"learning_rate": 1.9914448613738107e-05,
"loss": 0.9941,
"step": 125
},
{
"epoch": 0.14,
"learning_rate": 1.9887487036145942e-05,
"loss": 0.9852,
"step": 130
},
{
"epoch": 0.15,
"learning_rate": 1.985686165691514e-05,
"loss": 1.0073,
"step": 135
},
{
"epoch": 0.15,
"learning_rate": 1.9822583824259283e-05,
"loss": 1.0148,
"step": 140
},
{
"epoch": 0.16,
"learning_rate": 1.978466623980609e-05,
"loss": 0.9903,
"step": 145
},
{
"epoch": 0.17,
"learning_rate": 1.9743122953890857e-05,
"loss": 1.0055,
"step": 150
},
{
"epoch": 0.17,
"learning_rate": 1.9697969360350098e-05,
"loss": 0.9805,
"step": 155
},
{
"epoch": 0.18,
"learning_rate": 1.9649222190817382e-05,
"loss": 1.0283,
"step": 160
},
{
"epoch": 0.18,
"learning_rate": 1.9596899508523432e-05,
"loss": 1.0133,
"step": 165
},
{
"epoch": 0.19,
"learning_rate": 1.9541020701602814e-05,
"loss": 0.9921,
"step": 170
},
{
"epoch": 0.19,
"learning_rate": 1.948160647590966e-05,
"loss": 1.0006,
"step": 175
},
{
"epoch": 0.2,
"learning_rate": 1.941867884734515e-05,
"loss": 0.9872,
"step": 180
},
{
"epoch": 0.2,
"learning_rate": 1.9352261133699513e-05,
"loss": 1.0126,
"step": 185
},
{
"epoch": 0.21,
"learning_rate": 1.928237794601165e-05,
"loss": 1.0073,
"step": 190
},
{
"epoch": 0.21,
"learning_rate": 1.920905517944954e-05,
"loss": 1.0,
"step": 195
},
{
"epoch": 0.22,
"learning_rate": 1.9132320003714756e-05,
"loss": 1.0282,
"step": 200
},
{
"epoch": 0.23,
"learning_rate": 1.905220085297482e-05,
"loss": 1.0059,
"step": 205
},
{
"epoch": 0.23,
"learning_rate": 1.8968727415326885e-05,
"loss": 0.9929,
"step": 210
},
{
"epoch": 0.24,
"learning_rate": 1.888193062179685e-05,
"loss": 1.0006,
"step": 215
},
{
"epoch": 0.24,
"learning_rate": 1.87918426348779e-05,
"loss": 1.0212,
"step": 220
},
{
"epoch": 0.25,
"learning_rate": 1.8698496836612693e-05,
"loss": 1.0126,
"step": 225
},
{
"epoch": 0.25,
"learning_rate": 1.86019278162237e-05,
"loss": 0.9942,
"step": 230
},
{
"epoch": 0.26,
"learning_rate": 1.8502171357296144e-05,
"loss": 1.0087,
"step": 235
},
{
"epoch": 0.26,
"learning_rate": 1.8399264424518466e-05,
"loss": 0.9935,
"step": 240
},
{
"epoch": 0.27,
"learning_rate": 1.8293245149985055e-05,
"loss": 0.9924,
"step": 245
},
{
"epoch": 0.28,
"learning_rate": 1.8184152819066437e-05,
"loss": 0.9907,
"step": 250
},
{
"epoch": 0.28,
"learning_rate": 1.8072027855852098e-05,
"loss": 1.0176,
"step": 255
},
{
"epoch": 0.29,
"learning_rate": 1.7956911808171375e-05,
"loss": 0.9896,
"step": 260
},
{
"epoch": 0.29,
"learning_rate": 1.783884733219794e-05,
"loss": 1.0203,
"step": 265
},
{
"epoch": 0.3,
"learning_rate": 1.7717878176643563e-05,
"loss": 0.983,
"step": 270
},
{
"epoch": 0.3,
"learning_rate": 1.7594049166547073e-05,
"loss": 1.0278,
"step": 275
},
{
"epoch": 0.31,
"learning_rate": 1.7467406186664474e-05,
"loss": 1.0015,
"step": 280
},
{
"epoch": 0.31,
"learning_rate": 1.7337996164466373e-05,
"loss": 1.0147,
"step": 285
},
{
"epoch": 0.32,
"learning_rate": 1.7205867052749025e-05,
"loss": 0.989,
"step": 290
},
{
"epoch": 0.33,
"learning_rate": 1.7071067811865477e-05,
"loss": 0.9706,
"step": 295
},
{
"epoch": 0.33,
"learning_rate": 1.693364839158333e-05,
"loss": 0.9683,
"step": 300
},
{
"epoch": 0.34,
"learning_rate": 1.67936597125759e-05,
"loss": 0.9861,
"step": 305
},
{
"epoch": 0.34,
"learning_rate": 1.665115364755357e-05,
"loss": 1.0121,
"step": 310
},
{
"epoch": 0.35,
"learning_rate": 1.650618300204242e-05,
"loss": 1.0203,
"step": 315
},
{
"epoch": 0.35,
"learning_rate": 1.6358801494817172e-05,
"loss": 1.0106,
"step": 320
},
{
"epoch": 0.36,
"learning_rate": 1.6209063737995716e-05,
"loss": 1.0004,
"step": 325
},
{
"epoch": 0.36,
"learning_rate": 1.605702521680263e-05,
"loss": 0.994,
"step": 330
},
{
"epoch": 0.37,
"learning_rate": 1.5902742269009198e-05,
"loss": 1.0014,
"step": 335
},
{
"epoch": 0.37,
"learning_rate": 1.5746272064057438e-05,
"loss": 0.9998,
"step": 340
},
{
"epoch": 0.38,
"learning_rate": 1.5587672581876053e-05,
"loss": 0.9931,
"step": 345
},
{
"epoch": 0.39,
"learning_rate": 1.5427002591395966e-05,
"loss": 0.9998,
"step": 350
},
{
"epoch": 0.39,
"learning_rate": 1.526432162877356e-05,
"loss": 1.0229,
"step": 355
},
{
"epoch": 0.4,
"learning_rate": 1.5099689975329584e-05,
"loss": 0.998,
"step": 360
},
{
"epoch": 0.4,
"learning_rate": 1.4933168635211956e-05,
"loss": 0.9845,
"step": 365
},
{
"epoch": 0.41,
"learning_rate": 1.4764819312790706e-05,
"loss": 1.0018,
"step": 370
},
{
"epoch": 0.41,
"learning_rate": 1.4594704389793477e-05,
"loss": 0.9998,
"step": 375
},
{
"epoch": 0.42,
"learning_rate": 1.4422886902190014e-05,
"loss": 0.9824,
"step": 380
},
{
"epoch": 0.42,
"learning_rate": 1.4249430516834222e-05,
"loss": 1.0235,
"step": 385
},
{
"epoch": 0.43,
"learning_rate": 1.4074399507872455e-05,
"loss": 1.0178,
"step": 390
},
{
"epoch": 0.44,
"learning_rate": 1.3897858732926794e-05,
"loss": 0.9851,
"step": 395
},
{
"epoch": 0.44,
"learning_rate": 1.3719873609062078e-05,
"loss": 0.985,
"step": 400
},
{
"epoch": 0.45,
"learning_rate": 1.3540510088545652e-05,
"loss": 1.007,
"step": 405
},
{
"epoch": 0.45,
"learning_rate": 1.3359834634408828e-05,
"loss": 0.9794,
"step": 410
},
{
"epoch": 0.46,
"learning_rate": 1.3177914195819018e-05,
"loss": 1.0081,
"step": 415
},
{
"epoch": 0.46,
"learning_rate": 1.2994816183271787e-05,
"loss": 0.9973,
"step": 420
},
{
"epoch": 0.47,
"learning_rate": 1.2810608443611943e-05,
"loss": 1.0084,
"step": 425
},
{
"epoch": 0.47,
"learning_rate": 1.2625359234892906e-05,
"loss": 1.0051,
"step": 430
},
{
"epoch": 0.48,
"learning_rate": 1.2439137201083772e-05,
"loss": 1.0212,
"step": 435
},
{
"epoch": 0.49,
"learning_rate": 1.2252011346633306e-05,
"loss": 0.9887,
"step": 440
},
{
"epoch": 0.49,
"learning_rate": 1.2064051010900397e-05,
"loss": 0.9977,
"step": 445
},
{
"epoch": 0.5,
"learning_rate": 1.1875325842460424e-05,
"loss": 0.9808,
"step": 450
},
{
"epoch": 0.5,
"learning_rate": 1.1685905773296992e-05,
"loss": 0.9953,
"step": 455
},
{
"epoch": 0.51,
"learning_rate": 1.1495860992888712e-05,
"loss": 1.0066,
"step": 460
},
{
"epoch": 0.51,
"learning_rate": 1.130526192220052e-05,
"loss": 1.0059,
"step": 465
},
{
"epoch": 0.52,
"learning_rate": 1.1114179187589233e-05,
"loss": 1.0161,
"step": 470
},
{
"epoch": 0.52,
"learning_rate": 1.092268359463302e-05,
"loss": 0.9895,
"step": 475
},
{
"epoch": 0.53,
"learning_rate": 1.0730846101894428e-05,
"loss": 0.9961,
"step": 480
},
{
"epoch": 0.53,
"learning_rate": 1.0538737794626735e-05,
"loss": 1.0135,
"step": 485
},
{
"epoch": 0.54,
"learning_rate": 1.0346429858433354e-05,
"loss": 1.0086,
"step": 490
},
{
"epoch": 0.55,
"learning_rate": 1.0153993552890069e-05,
"loss": 0.9978,
"step": 495
},
{
"epoch": 0.55,
"learning_rate": 9.961500185139843e-06,
"loss": 0.9962,
"step": 500
},
{
"epoch": 0.56,
"learning_rate": 9.769021083469991e-06,
"loss": 0.9872,
"step": 505
},
{
"epoch": 0.56,
"learning_rate": 9.576627570881551e-06,
"loss": 0.9868,
"step": 510
},
{
"epoch": 0.57,
"learning_rate": 9.384390938660572e-06,
"loss": 1.0103,
"step": 515
},
{
"epoch": 0.57,
"learning_rate": 9.192382419961208e-06,
"loss": 0.9979,
"step": 520
},
{
"epoch": 0.58,
"learning_rate": 9.000673163410287e-06,
"loss": 0.9763,
"step": 525
},
{
"epoch": 0.58,
"learning_rate": 8.809334206743251e-06,
"loss": 0.9916,
"step": 530
},
{
"epoch": 0.59,
"learning_rate": 8.618436450481182e-06,
"loss": 0.991,
"step": 535
},
{
"epoch": 0.6,
"learning_rate": 8.428050631658628e-06,
"loss": 1.0002,
"step": 540
},
{
"epoch": 0.6,
"learning_rate": 8.238247297612092e-06,
"loss": 0.9959,
"step": 545
},
{
"epoch": 0.61,
"learning_rate": 8.04909677983872e-06,
"loss": 0.97,
"step": 550
},
{
"epoch": 0.61,
"learning_rate": 7.860669167935028e-06,
"loss": 0.9843,
"step": 555
},
{
"epoch": 0.62,
"learning_rate": 7.673034283625258e-06,
"loss": 0.985,
"step": 560
},
{
"epoch": 0.62,
"learning_rate": 7.486261654888974e-06,
"loss": 1.0089,
"step": 565
},
{
"epoch": 0.63,
"learning_rate": 7.300420490197524e-06,
"loss": 1.0078,
"step": 570
},
{
"epoch": 0.63,
"learning_rate": 7.115579652868878e-06,
"loss": 0.9974,
"step": 575
},
{
"epoch": 0.64,
"learning_rate": 6.9318076355503835e-06,
"loss": 1.0002,
"step": 580
},
{
"epoch": 0.64,
"learning_rate": 6.749172534838849e-06,
"loss": 0.989,
"step": 585
},
{
"epoch": 0.65,
"learning_rate": 6.5677420260474055e-06,
"loss": 0.9956,
"step": 590
},
{
"epoch": 0.66,
"learning_rate": 6.387583338128471e-06,
"loss": 0.9893,
"step": 595
},
{
"epoch": 0.66,
"learning_rate": 6.208763228762082e-06,
"loss": 1.0192,
"step": 600
},
{
"epoch": 0.67,
"learning_rate": 6.031347959618914e-06,
"loss": 0.9641,
"step": 605
},
{
"epoch": 0.67,
"learning_rate": 5.855403271807033e-06,
"loss": 0.9919,
"step": 610
},
{
"epoch": 0.68,
"learning_rate": 5.6809943615115915e-06,
"loss": 1.0092,
"step": 615
},
{
"epoch": 0.68,
"learning_rate": 5.508185855836426e-06,
"loss": 1.0071,
"step": 620
},
{
"epoch": 0.69,
"learning_rate": 5.337041788856518e-06,
"loss": 1.0032,
"step": 625
},
{
"epoch": 0.69,
"learning_rate": 5.1676255778902225e-06,
"loss": 1.0175,
"step": 630
},
{
"epoch": 0.7,
"learning_rate": 5.000000000000003e-06,
"loss": 1.0074,
"step": 635
},
{
"epoch": 0.71,
"learning_rate": 4.834227168730451e-06,
"loss": 1.0327,
"step": 640
},
{
"epoch": 0.71,
"learning_rate": 4.670368511092166e-06,
"loss": 1.0093,
"step": 645
},
{
"epoch": 0.72,
"learning_rate": 4.508484744799979e-06,
"loss": 1.0004,
"step": 650
},
{
"epoch": 0.72,
"learning_rate": 4.348635855774082e-06,
"loss": 0.9893,
"step": 655
},
{
"epoch": 0.73,
"learning_rate": 4.190881075912269e-06,
"loss": 0.9921,
"step": 660
},
{
"epoch": 0.73,
"learning_rate": 4.035278861141588e-06,
"loss": 1.0021,
"step": 665
},
{
"epoch": 0.74,
"learning_rate": 3.881886869757565e-06,
"loss": 1.0096,
"step": 670
},
{
"epoch": 0.74,
"learning_rate": 3.730761941058938e-06,
"loss": 0.9959,
"step": 675
},
{
"epoch": 0.75,
"learning_rate": 3.581960074285885e-06,
"loss": 0.9875,
"step": 680
},
{
"epoch": 0.76,
"learning_rate": 3.435536407869575e-06,
"loss": 0.9872,
"step": 685
},
{
"epoch": 0.76,
"learning_rate": 3.2915451990006365e-06,
"loss": 0.9988,
"step": 690
},
{
"epoch": 0.77,
"learning_rate": 3.150039803524194e-06,
"loss": 0.9843,
"step": 695
},
{
"epoch": 0.77,
"learning_rate": 3.011072656168906e-06,
"loss": 0.9997,
"step": 700
},
{
"epoch": 0.78,
"learning_rate": 2.874695251117303e-06,
"loss": 0.9938,
"step": 705
},
{
"epoch": 0.78,
"learning_rate": 2.7409581229246496e-06,
"loss": 0.9863,
"step": 710
},
{
"epoch": 0.79,
"learning_rate": 2.6099108277934105e-06,
"loss": 1.0175,
"step": 715
},
{
"epoch": 0.79,
"learning_rate": 2.4816019252102274e-06,
"loss": 0.9998,
"step": 720
},
{
"epoch": 0.8,
"learning_rate": 2.3560789599522325e-06,
"loss": 0.9945,
"step": 725
},
{
"epoch": 0.8,
"learning_rate": 2.2333884444693656e-06,
"loss": 0.99,
"step": 730
},
{
"epoch": 0.81,
"learning_rate": 2.1135758416492168e-06,
"loss": 0.9848,
"step": 735
},
{
"epoch": 0.82,
"learning_rate": 1.996685547970787e-06,
"loss": 0.9955,
"step": 740
},
{
"epoch": 0.82,
"learning_rate": 1.882760877053388e-06,
"loss": 0.9842,
"step": 745
},
{
"epoch": 0.83,
"learning_rate": 1.7718440436068385e-06,
"loss": 1.0163,
"step": 750
},
{
"epoch": 0.83,
"learning_rate": 1.663976147788806e-06,
"loss": 0.9938,
"step": 755
},
{
"epoch": 0.84,
"learning_rate": 1.55919715997518e-06,
"loss": 1.0167,
"step": 760
},
{
"epoch": 0.84,
"learning_rate": 1.4575459059490771e-06,
"loss": 1.0017,
"step": 765
},
{
"epoch": 0.85,
"learning_rate": 1.3590600525139764e-06,
"loss": 0.9964,
"step": 770
},
{
"epoch": 0.85,
"learning_rate": 1.2637760935363053e-06,
"loss": 1.0156,
"step": 775
},
{
"epoch": 0.86,
"learning_rate": 1.1717293364226612e-06,
"loss": 1.0111,
"step": 780
},
{
"epoch": 0.87,
"learning_rate": 1.0829538890366864e-06,
"loss": 1.007,
"step": 785
},
{
"epoch": 0.87,
"learning_rate": 9.97482647060405e-07,
"loss": 0.9782,
"step": 790
},
{
"epoch": 0.88,
"learning_rate": 9.153472818047627e-07,
"loss": 1.012,
"step": 795
},
{
"epoch": 0.88,
"learning_rate": 8.365782284738488e-07,
"loss": 0.9846,
"step": 800
},
{
"epoch": 0.89,
"learning_rate": 7.612046748871327e-07,
"loss": 1.0234,
"step": 805
},
{
"epoch": 0.89,
"learning_rate": 6.892545506639559e-07,
"loss": 1.0155,
"step": 810
},
{
"epoch": 0.9,
"learning_rate": 6.207545168742212e-07,
"loss": 1.0099,
"step": 815
},
{
"epoch": 0.9,
"learning_rate": 5.557299561591478e-07,
"loss": 0.982,
"step": 820
},
{
"epoch": 0.91,
"learning_rate": 4.942049633257673e-07,
"loss": 1.0155,
"step": 825
},
{
"epoch": 0.92,
"learning_rate": 4.3620233641858855e-07,
"loss": 1.0128,
"step": 830
},
{
"epoch": 0.92,
"learning_rate": 3.817435682718096e-07,
"loss": 0.9917,
"step": 835
},
{
"epoch": 0.93,
"learning_rate": 3.3084883854516136e-07,
"loss": 1.0075,
"step": 840
},
{
"epoch": 0.93,
"learning_rate": 2.835370062463449e-07,
"loss": 0.9815,
"step": 845
},
{
"epoch": 0.94,
"learning_rate": 2.398256027428436e-07,
"loss": 0.9992,
"step": 850
},
{
"epoch": 0.94,
"learning_rate": 1.9973082526568155e-07,
"loss": 1.0018,
"step": 855
},
{
"epoch": 0.95,
"learning_rate": 1.6326753090755487e-07,
"loss": 1.0033,
"step": 860
},
{
"epoch": 0.95,
"learning_rate": 1.3044923111754427e-07,
"loss": 0.9775,
"step": 865
},
{
"epoch": 0.96,
"learning_rate": 1.0128808669445323e-07,
"loss": 0.9933,
"step": 870
},
{
"epoch": 0.96,
"learning_rate": 7.579490328064265e-08,
"loss": 0.9935,
"step": 875
},
{
"epoch": 0.97,
"learning_rate": 5.397912735799371e-08,
"loss": 0.9873,
"step": 880
},
{
"epoch": 0.98,
"learning_rate": 3.5848842747532306e-08,
"loss": 0.9878,
"step": 885
},
{
"epoch": 0.98,
"learning_rate": 2.1410767613965212e-08,
"loss": 1.0115,
"step": 890
},
{
"epoch": 0.99,
"learning_rate": 1.0670251976275803e-08,
"loss": 1.0062,
"step": 895
},
{
"epoch": 0.99,
"learning_rate": 3.63127572527433e-09,
"loss": 0.9915,
"step": 900
},
{
"epoch": 1.0,
"learning_rate": 2.964471488531295e-10,
"loss": 1.014,
"step": 905
},
{
"epoch": 1.0,
"eval_loss": 1.004878044128418,
"eval_runtime": 471.2518,
"eval_samples_per_second": 13.666,
"eval_steps_per_second": 0.855,
"step": 907
},
{
"epoch": 1.0,
"step": 907,
"total_flos": 5.321288023160127e+18,
"train_loss": 1.0006486285219118,
"train_runtime": 14584.0452,
"train_samples_per_second": 3.979,
"train_steps_per_second": 0.062
}
],
"logging_steps": 5,
"max_steps": 907,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 100,
"total_flos": 5.321288023160127e+18,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}