kbressem
updating weights after training with data v3
306511f
{
"best_metric": 1.6129449605941772,
"best_model_checkpoint": "./lora-alpaca-7b/checkpoint-3400",
"epoch": 4.644808743169399,
"global_step": 3400,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 1.8000000000000001e-06,
"loss": 2.4672,
"step": 10
},
{
"epoch": 0.03,
"learning_rate": 3.6000000000000003e-06,
"loss": 2.4812,
"step": 20
},
{
"epoch": 0.04,
"learning_rate": 5.600000000000001e-06,
"loss": 2.5302,
"step": 30
},
{
"epoch": 0.05,
"learning_rate": 7.4e-06,
"loss": 2.4524,
"step": 40
},
{
"epoch": 0.07,
"learning_rate": 9.4e-06,
"loss": 2.5016,
"step": 50
},
{
"epoch": 0.08,
"learning_rate": 1.14e-05,
"loss": 2.4367,
"step": 60
},
{
"epoch": 0.1,
"learning_rate": 1.3200000000000002e-05,
"loss": 2.4187,
"step": 70
},
{
"epoch": 0.11,
"learning_rate": 1.5200000000000002e-05,
"loss": 2.4319,
"step": 80
},
{
"epoch": 0.12,
"learning_rate": 1.72e-05,
"loss": 2.327,
"step": 90
},
{
"epoch": 0.14,
"learning_rate": 1.9200000000000003e-05,
"loss": 2.2739,
"step": 100
},
{
"epoch": 0.15,
"learning_rate": 1.996629213483146e-05,
"loss": 2.1875,
"step": 110
},
{
"epoch": 0.16,
"learning_rate": 1.9910112359550565e-05,
"loss": 2.0791,
"step": 120
},
{
"epoch": 0.18,
"learning_rate": 1.9853932584269663e-05,
"loss": 2.0469,
"step": 130
},
{
"epoch": 0.19,
"learning_rate": 1.9797752808988765e-05,
"loss": 1.9487,
"step": 140
},
{
"epoch": 0.2,
"learning_rate": 1.9741573033707866e-05,
"loss": 1.8747,
"step": 150
},
{
"epoch": 0.22,
"learning_rate": 1.9685393258426968e-05,
"loss": 1.8928,
"step": 160
},
{
"epoch": 0.23,
"learning_rate": 1.962921348314607e-05,
"loss": 1.8172,
"step": 170
},
{
"epoch": 0.25,
"learning_rate": 1.957303370786517e-05,
"loss": 1.8132,
"step": 180
},
{
"epoch": 0.26,
"learning_rate": 1.9516853932584272e-05,
"loss": 1.7915,
"step": 190
},
{
"epoch": 0.27,
"learning_rate": 1.9460674157303373e-05,
"loss": 1.7315,
"step": 200
},
{
"epoch": 0.27,
"eval_loss": 1.8124021291732788,
"eval_runtime": 16.7109,
"eval_samples_per_second": 119.683,
"eval_steps_per_second": 1.915,
"step": 200
},
{
"epoch": 0.29,
"learning_rate": 1.9404494382022475e-05,
"loss": 1.8106,
"step": 210
},
{
"epoch": 0.3,
"learning_rate": 1.9348314606741573e-05,
"loss": 1.7569,
"step": 220
},
{
"epoch": 0.31,
"learning_rate": 1.9292134831460678e-05,
"loss": 1.7769,
"step": 230
},
{
"epoch": 0.33,
"learning_rate": 1.9235955056179776e-05,
"loss": 1.7745,
"step": 240
},
{
"epoch": 0.34,
"learning_rate": 1.9179775280898877e-05,
"loss": 1.712,
"step": 250
},
{
"epoch": 0.36,
"learning_rate": 1.912359550561798e-05,
"loss": 1.804,
"step": 260
},
{
"epoch": 0.37,
"learning_rate": 1.906741573033708e-05,
"loss": 1.7397,
"step": 270
},
{
"epoch": 0.38,
"learning_rate": 1.901123595505618e-05,
"loss": 1.7582,
"step": 280
},
{
"epoch": 0.4,
"learning_rate": 1.8955056179775283e-05,
"loss": 1.7604,
"step": 290
},
{
"epoch": 0.41,
"learning_rate": 1.8898876404494384e-05,
"loss": 1.6978,
"step": 300
},
{
"epoch": 0.42,
"learning_rate": 1.8842696629213482e-05,
"loss": 1.7822,
"step": 310
},
{
"epoch": 0.44,
"learning_rate": 1.8786516853932587e-05,
"loss": 1.7078,
"step": 320
},
{
"epoch": 0.45,
"learning_rate": 1.8730337078651685e-05,
"loss": 1.6948,
"step": 330
},
{
"epoch": 0.46,
"learning_rate": 1.8674157303370787e-05,
"loss": 1.6784,
"step": 340
},
{
"epoch": 0.48,
"learning_rate": 1.861797752808989e-05,
"loss": 1.5832,
"step": 350
},
{
"epoch": 0.49,
"learning_rate": 1.856179775280899e-05,
"loss": 1.6874,
"step": 360
},
{
"epoch": 0.51,
"learning_rate": 1.850561797752809e-05,
"loss": 1.6197,
"step": 370
},
{
"epoch": 0.52,
"learning_rate": 1.8449438202247192e-05,
"loss": 1.6153,
"step": 380
},
{
"epoch": 0.53,
"learning_rate": 1.8393258426966294e-05,
"loss": 1.6466,
"step": 390
},
{
"epoch": 0.55,
"learning_rate": 1.8337078651685395e-05,
"loss": 1.5556,
"step": 400
},
{
"epoch": 0.55,
"eval_loss": 1.6927136182785034,
"eval_runtime": 16.7427,
"eval_samples_per_second": 119.455,
"eval_steps_per_second": 1.911,
"step": 400
},
{
"epoch": 0.56,
"learning_rate": 1.8280898876404497e-05,
"loss": 1.6737,
"step": 410
},
{
"epoch": 0.57,
"learning_rate": 1.8224719101123598e-05,
"loss": 1.6131,
"step": 420
},
{
"epoch": 0.59,
"learning_rate": 1.81685393258427e-05,
"loss": 1.6083,
"step": 430
},
{
"epoch": 0.6,
"learning_rate": 1.81123595505618e-05,
"loss": 1.6308,
"step": 440
},
{
"epoch": 0.61,
"learning_rate": 1.80561797752809e-05,
"loss": 1.5515,
"step": 450
},
{
"epoch": 0.63,
"learning_rate": 1.8e-05,
"loss": 1.6711,
"step": 460
},
{
"epoch": 0.64,
"learning_rate": 1.7943820224719102e-05,
"loss": 1.6046,
"step": 470
},
{
"epoch": 0.66,
"learning_rate": 1.7887640449438203e-05,
"loss": 1.6142,
"step": 480
},
{
"epoch": 0.67,
"learning_rate": 1.7831460674157305e-05,
"loss": 1.629,
"step": 490
},
{
"epoch": 0.68,
"learning_rate": 1.7775280898876406e-05,
"loss": 1.5448,
"step": 500
},
{
"epoch": 0.7,
"learning_rate": 1.7719101123595508e-05,
"loss": 1.6575,
"step": 510
},
{
"epoch": 0.71,
"learning_rate": 1.766292134831461e-05,
"loss": 1.6023,
"step": 520
},
{
"epoch": 0.72,
"learning_rate": 1.760674157303371e-05,
"loss": 1.6008,
"step": 530
},
{
"epoch": 0.74,
"learning_rate": 1.755056179775281e-05,
"loss": 1.6152,
"step": 540
},
{
"epoch": 0.75,
"learning_rate": 1.7494382022471913e-05,
"loss": 1.5349,
"step": 550
},
{
"epoch": 0.77,
"learning_rate": 1.743820224719101e-05,
"loss": 1.6596,
"step": 560
},
{
"epoch": 0.78,
"learning_rate": 1.7382022471910113e-05,
"loss": 1.5952,
"step": 570
},
{
"epoch": 0.79,
"learning_rate": 1.7325842696629214e-05,
"loss": 1.5923,
"step": 580
},
{
"epoch": 0.81,
"learning_rate": 1.7269662921348316e-05,
"loss": 1.612,
"step": 590
},
{
"epoch": 0.82,
"learning_rate": 1.7213483146067417e-05,
"loss": 1.5331,
"step": 600
},
{
"epoch": 0.82,
"eval_loss": 1.6660429239273071,
"eval_runtime": 16.7438,
"eval_samples_per_second": 119.447,
"eval_steps_per_second": 1.911,
"step": 600
},
{
"epoch": 0.83,
"learning_rate": 1.715730337078652e-05,
"loss": 1.66,
"step": 610
},
{
"epoch": 0.85,
"learning_rate": 1.710112359550562e-05,
"loss": 1.5926,
"step": 620
},
{
"epoch": 0.86,
"learning_rate": 1.7044943820224718e-05,
"loss": 1.5971,
"step": 630
},
{
"epoch": 0.87,
"learning_rate": 1.6988764044943823e-05,
"loss": 1.6151,
"step": 640
},
{
"epoch": 0.89,
"learning_rate": 1.693258426966292e-05,
"loss": 1.5259,
"step": 650
},
{
"epoch": 0.9,
"learning_rate": 1.6876404494382022e-05,
"loss": 1.6578,
"step": 660
},
{
"epoch": 0.92,
"learning_rate": 1.6820224719101127e-05,
"loss": 1.5869,
"step": 670
},
{
"epoch": 0.93,
"learning_rate": 1.6764044943820225e-05,
"loss": 1.5802,
"step": 680
},
{
"epoch": 0.94,
"learning_rate": 1.6707865168539327e-05,
"loss": 1.6185,
"step": 690
},
{
"epoch": 0.96,
"learning_rate": 1.6651685393258428e-05,
"loss": 1.5177,
"step": 700
},
{
"epoch": 0.97,
"learning_rate": 1.659550561797753e-05,
"loss": 1.6455,
"step": 710
},
{
"epoch": 0.98,
"learning_rate": 1.653932584269663e-05,
"loss": 1.5986,
"step": 720
},
{
"epoch": 1.0,
"learning_rate": 1.6483146067415732e-05,
"loss": 1.5889,
"step": 730
},
{
"epoch": 1.01,
"learning_rate": 1.6426966292134834e-05,
"loss": 1.5716,
"step": 740
},
{
"epoch": 1.02,
"learning_rate": 1.6370786516853935e-05,
"loss": 1.6027,
"step": 750
},
{
"epoch": 1.04,
"learning_rate": 1.6314606741573037e-05,
"loss": 1.6054,
"step": 760
},
{
"epoch": 1.05,
"learning_rate": 1.6258426966292135e-05,
"loss": 1.5594,
"step": 770
},
{
"epoch": 1.07,
"learning_rate": 1.6202247191011236e-05,
"loss": 1.643,
"step": 780
},
{
"epoch": 1.08,
"learning_rate": 1.6146067415730338e-05,
"loss": 1.5481,
"step": 790
},
{
"epoch": 1.09,
"learning_rate": 1.608988764044944e-05,
"loss": 1.6022,
"step": 800
},
{
"epoch": 1.09,
"eval_loss": 1.65511155128479,
"eval_runtime": 16.733,
"eval_samples_per_second": 119.524,
"eval_steps_per_second": 1.912,
"step": 800
},
{
"epoch": 1.11,
"learning_rate": 1.603370786516854e-05,
"loss": 1.5974,
"step": 810
},
{
"epoch": 1.12,
"learning_rate": 1.5977528089887642e-05,
"loss": 1.55,
"step": 820
},
{
"epoch": 1.13,
"learning_rate": 1.5921348314606743e-05,
"loss": 1.6455,
"step": 830
},
{
"epoch": 1.15,
"learning_rate": 1.5865168539325845e-05,
"loss": 1.5453,
"step": 840
},
{
"epoch": 1.16,
"learning_rate": 1.5808988764044946e-05,
"loss": 1.5995,
"step": 850
},
{
"epoch": 1.17,
"learning_rate": 1.5752808988764044e-05,
"loss": 1.5855,
"step": 860
},
{
"epoch": 1.19,
"learning_rate": 1.569662921348315e-05,
"loss": 1.555,
"step": 870
},
{
"epoch": 1.2,
"learning_rate": 1.5640449438202247e-05,
"loss": 1.6454,
"step": 880
},
{
"epoch": 1.22,
"learning_rate": 1.558426966292135e-05,
"loss": 1.5501,
"step": 890
},
{
"epoch": 1.23,
"learning_rate": 1.5528089887640453e-05,
"loss": 1.6016,
"step": 900
},
{
"epoch": 1.24,
"learning_rate": 1.547191011235955e-05,
"loss": 1.5892,
"step": 910
},
{
"epoch": 1.26,
"learning_rate": 1.5415730337078653e-05,
"loss": 1.5466,
"step": 920
},
{
"epoch": 1.27,
"learning_rate": 1.5359550561797754e-05,
"loss": 1.6446,
"step": 930
},
{
"epoch": 1.28,
"learning_rate": 1.5303370786516856e-05,
"loss": 1.5502,
"step": 940
},
{
"epoch": 1.3,
"learning_rate": 1.5247191011235955e-05,
"loss": 1.5983,
"step": 950
},
{
"epoch": 1.31,
"learning_rate": 1.5191011235955057e-05,
"loss": 1.5972,
"step": 960
},
{
"epoch": 1.33,
"learning_rate": 1.5134831460674158e-05,
"loss": 1.5438,
"step": 970
},
{
"epoch": 1.34,
"learning_rate": 1.507865168539326e-05,
"loss": 1.6339,
"step": 980
},
{
"epoch": 1.35,
"learning_rate": 1.5022471910112361e-05,
"loss": 1.5421,
"step": 990
},
{
"epoch": 1.37,
"learning_rate": 1.4966292134831461e-05,
"loss": 1.5728,
"step": 1000
},
{
"epoch": 1.37,
"eval_loss": 1.6474125385284424,
"eval_runtime": 16.7383,
"eval_samples_per_second": 119.486,
"eval_steps_per_second": 1.912,
"step": 1000
},
{
"epoch": 1.38,
"learning_rate": 1.4910112359550564e-05,
"loss": 1.5847,
"step": 1010
},
{
"epoch": 1.39,
"learning_rate": 1.4853932584269664e-05,
"loss": 1.5442,
"step": 1020
},
{
"epoch": 1.41,
"learning_rate": 1.4797752808988765e-05,
"loss": 1.6359,
"step": 1030
},
{
"epoch": 1.42,
"learning_rate": 1.4741573033707865e-05,
"loss": 1.535,
"step": 1040
},
{
"epoch": 1.43,
"learning_rate": 1.4685393258426968e-05,
"loss": 1.5923,
"step": 1050
},
{
"epoch": 1.45,
"learning_rate": 1.462921348314607e-05,
"loss": 1.5966,
"step": 1060
},
{
"epoch": 1.46,
"learning_rate": 1.457303370786517e-05,
"loss": 1.5491,
"step": 1070
},
{
"epoch": 1.48,
"learning_rate": 1.4516853932584272e-05,
"loss": 1.6308,
"step": 1080
},
{
"epoch": 1.49,
"learning_rate": 1.4460674157303372e-05,
"loss": 1.5383,
"step": 1090
},
{
"epoch": 1.5,
"learning_rate": 1.4404494382022474e-05,
"loss": 1.5848,
"step": 1100
},
{
"epoch": 1.52,
"learning_rate": 1.4348314606741573e-05,
"loss": 1.5846,
"step": 1110
},
{
"epoch": 1.53,
"learning_rate": 1.4292134831460676e-05,
"loss": 1.5461,
"step": 1120
},
{
"epoch": 1.54,
"learning_rate": 1.4235955056179776e-05,
"loss": 1.6245,
"step": 1130
},
{
"epoch": 1.56,
"learning_rate": 1.4179775280898878e-05,
"loss": 1.527,
"step": 1140
},
{
"epoch": 1.57,
"learning_rate": 1.4123595505617979e-05,
"loss": 1.5902,
"step": 1150
},
{
"epoch": 1.58,
"learning_rate": 1.4067415730337079e-05,
"loss": 1.5879,
"step": 1160
},
{
"epoch": 1.6,
"learning_rate": 1.4011235955056182e-05,
"loss": 1.5396,
"step": 1170
},
{
"epoch": 1.61,
"learning_rate": 1.3955056179775282e-05,
"loss": 1.6325,
"step": 1180
},
{
"epoch": 1.63,
"learning_rate": 1.3898876404494383e-05,
"loss": 1.5381,
"step": 1190
},
{
"epoch": 1.64,
"learning_rate": 1.3842696629213483e-05,
"loss": 1.5918,
"step": 1200
},
{
"epoch": 1.64,
"eval_loss": 1.6412869691848755,
"eval_runtime": 16.7579,
"eval_samples_per_second": 119.347,
"eval_steps_per_second": 1.91,
"step": 1200
},
{
"epoch": 1.65,
"learning_rate": 1.3786516853932586e-05,
"loss": 1.5861,
"step": 1210
},
{
"epoch": 1.67,
"learning_rate": 1.3730337078651686e-05,
"loss": 1.5399,
"step": 1220
},
{
"epoch": 1.68,
"learning_rate": 1.3674157303370787e-05,
"loss": 1.6292,
"step": 1230
},
{
"epoch": 1.69,
"learning_rate": 1.361797752808989e-05,
"loss": 1.5349,
"step": 1240
},
{
"epoch": 1.71,
"learning_rate": 1.356179775280899e-05,
"loss": 1.5818,
"step": 1250
},
{
"epoch": 1.72,
"learning_rate": 1.3505617977528091e-05,
"loss": 1.5689,
"step": 1260
},
{
"epoch": 1.73,
"learning_rate": 1.3449438202247191e-05,
"loss": 1.5437,
"step": 1270
},
{
"epoch": 1.75,
"learning_rate": 1.3393258426966294e-05,
"loss": 1.6323,
"step": 1280
},
{
"epoch": 1.76,
"learning_rate": 1.3337078651685394e-05,
"loss": 1.5402,
"step": 1290
},
{
"epoch": 1.78,
"learning_rate": 1.3280898876404495e-05,
"loss": 1.5898,
"step": 1300
},
{
"epoch": 1.79,
"learning_rate": 1.3224719101123597e-05,
"loss": 1.5838,
"step": 1310
},
{
"epoch": 1.8,
"learning_rate": 1.3168539325842698e-05,
"loss": 1.5306,
"step": 1320
},
{
"epoch": 1.82,
"learning_rate": 1.31123595505618e-05,
"loss": 1.6181,
"step": 1330
},
{
"epoch": 1.83,
"learning_rate": 1.30561797752809e-05,
"loss": 1.5343,
"step": 1340
},
{
"epoch": 1.84,
"learning_rate": 1.3000000000000001e-05,
"loss": 1.5789,
"step": 1350
},
{
"epoch": 1.86,
"learning_rate": 1.29438202247191e-05,
"loss": 1.5799,
"step": 1360
},
{
"epoch": 1.87,
"learning_rate": 1.2887640449438204e-05,
"loss": 1.543,
"step": 1370
},
{
"epoch": 1.89,
"learning_rate": 1.2831460674157303e-05,
"loss": 1.6282,
"step": 1380
},
{
"epoch": 1.9,
"learning_rate": 1.2775280898876405e-05,
"loss": 1.526,
"step": 1390
},
{
"epoch": 1.91,
"learning_rate": 1.2719101123595508e-05,
"loss": 1.5751,
"step": 1400
},
{
"epoch": 1.91,
"eval_loss": 1.635980248451233,
"eval_runtime": 16.7435,
"eval_samples_per_second": 119.449,
"eval_steps_per_second": 1.911,
"step": 1400
},
{
"epoch": 1.93,
"learning_rate": 1.2662921348314608e-05,
"loss": 1.5793,
"step": 1410
},
{
"epoch": 1.94,
"learning_rate": 1.260674157303371e-05,
"loss": 1.5381,
"step": 1420
},
{
"epoch": 1.95,
"learning_rate": 1.2550561797752809e-05,
"loss": 1.6373,
"step": 1430
},
{
"epoch": 1.97,
"learning_rate": 1.2494382022471912e-05,
"loss": 1.5289,
"step": 1440
},
{
"epoch": 1.98,
"learning_rate": 1.2438202247191012e-05,
"loss": 1.5903,
"step": 1450
},
{
"epoch": 1.99,
"learning_rate": 1.2382022471910113e-05,
"loss": 1.583,
"step": 1460
},
{
"epoch": 2.01,
"learning_rate": 1.2325842696629215e-05,
"loss": 1.5091,
"step": 1470
},
{
"epoch": 2.02,
"learning_rate": 1.2269662921348316e-05,
"loss": 1.6329,
"step": 1480
},
{
"epoch": 2.04,
"learning_rate": 1.2213483146067418e-05,
"loss": 1.5245,
"step": 1490
},
{
"epoch": 2.05,
"learning_rate": 1.2157303370786517e-05,
"loss": 1.5883,
"step": 1500
},
{
"epoch": 2.06,
"learning_rate": 1.2101123595505619e-05,
"loss": 1.5845,
"step": 1510
},
{
"epoch": 2.08,
"learning_rate": 1.2044943820224718e-05,
"loss": 1.4885,
"step": 1520
},
{
"epoch": 2.09,
"learning_rate": 1.1988764044943822e-05,
"loss": 1.6478,
"step": 1530
},
{
"epoch": 2.1,
"learning_rate": 1.1932584269662921e-05,
"loss": 1.5219,
"step": 1540
},
{
"epoch": 2.12,
"learning_rate": 1.1876404494382023e-05,
"loss": 1.5926,
"step": 1550
},
{
"epoch": 2.13,
"learning_rate": 1.1820224719101126e-05,
"loss": 1.5909,
"step": 1560
},
{
"epoch": 2.14,
"learning_rate": 1.1764044943820226e-05,
"loss": 1.4917,
"step": 1570
},
{
"epoch": 2.16,
"learning_rate": 1.1707865168539327e-05,
"loss": 1.6298,
"step": 1580
},
{
"epoch": 2.17,
"learning_rate": 1.1651685393258427e-05,
"loss": 1.5156,
"step": 1590
},
{
"epoch": 2.19,
"learning_rate": 1.159550561797753e-05,
"loss": 1.5782,
"step": 1600
},
{
"epoch": 2.19,
"eval_loss": 1.631454348564148,
"eval_runtime": 16.7278,
"eval_samples_per_second": 119.561,
"eval_steps_per_second": 1.913,
"step": 1600
},
{
"epoch": 2.2,
"learning_rate": 1.153932584269663e-05,
"loss": 1.5834,
"step": 1610
},
{
"epoch": 2.21,
"learning_rate": 1.1483146067415731e-05,
"loss": 1.4866,
"step": 1620
},
{
"epoch": 2.23,
"learning_rate": 1.1426966292134834e-05,
"loss": 1.6419,
"step": 1630
},
{
"epoch": 2.24,
"learning_rate": 1.1370786516853934e-05,
"loss": 1.5204,
"step": 1640
},
{
"epoch": 2.25,
"learning_rate": 1.1314606741573035e-05,
"loss": 1.5896,
"step": 1650
},
{
"epoch": 2.27,
"learning_rate": 1.1258426966292135e-05,
"loss": 1.5813,
"step": 1660
},
{
"epoch": 2.28,
"learning_rate": 1.1202247191011237e-05,
"loss": 1.4843,
"step": 1670
},
{
"epoch": 2.3,
"learning_rate": 1.1146067415730338e-05,
"loss": 1.6408,
"step": 1680
},
{
"epoch": 2.31,
"learning_rate": 1.108988764044944e-05,
"loss": 1.516,
"step": 1690
},
{
"epoch": 2.32,
"learning_rate": 1.103370786516854e-05,
"loss": 1.5969,
"step": 1700
},
{
"epoch": 2.34,
"learning_rate": 1.097752808988764e-05,
"loss": 1.5933,
"step": 1710
},
{
"epoch": 2.35,
"learning_rate": 1.0921348314606744e-05,
"loss": 1.4849,
"step": 1720
},
{
"epoch": 2.36,
"learning_rate": 1.0865168539325843e-05,
"loss": 1.6265,
"step": 1730
},
{
"epoch": 2.38,
"learning_rate": 1.0808988764044945e-05,
"loss": 1.5128,
"step": 1740
},
{
"epoch": 2.39,
"learning_rate": 1.0752808988764045e-05,
"loss": 1.5925,
"step": 1750
},
{
"epoch": 2.4,
"learning_rate": 1.0696629213483148e-05,
"loss": 1.585,
"step": 1760
},
{
"epoch": 2.42,
"learning_rate": 1.0640449438202248e-05,
"loss": 1.4848,
"step": 1770
},
{
"epoch": 2.43,
"learning_rate": 1.0584269662921349e-05,
"loss": 1.6436,
"step": 1780
},
{
"epoch": 2.45,
"learning_rate": 1.0528089887640452e-05,
"loss": 1.5123,
"step": 1790
},
{
"epoch": 2.46,
"learning_rate": 1.0471910112359552e-05,
"loss": 1.5921,
"step": 1800
},
{
"epoch": 2.46,
"eval_loss": 1.6276172399520874,
"eval_runtime": 16.7327,
"eval_samples_per_second": 119.527,
"eval_steps_per_second": 1.912,
"step": 1800
},
{
"epoch": 2.47,
"learning_rate": 1.0415730337078653e-05,
"loss": 1.5806,
"step": 1810
},
{
"epoch": 2.49,
"learning_rate": 1.0359550561797753e-05,
"loss": 1.4778,
"step": 1820
},
{
"epoch": 2.5,
"learning_rate": 1.0303370786516856e-05,
"loss": 1.6308,
"step": 1830
},
{
"epoch": 2.51,
"learning_rate": 1.0247191011235956e-05,
"loss": 1.5083,
"step": 1840
},
{
"epoch": 2.53,
"learning_rate": 1.0191011235955057e-05,
"loss": 1.5761,
"step": 1850
},
{
"epoch": 2.54,
"learning_rate": 1.0134831460674157e-05,
"loss": 1.5833,
"step": 1860
},
{
"epoch": 2.55,
"learning_rate": 1.0078651685393258e-05,
"loss": 1.483,
"step": 1870
},
{
"epoch": 2.57,
"learning_rate": 1.0022471910112362e-05,
"loss": 1.6321,
"step": 1880
},
{
"epoch": 2.58,
"learning_rate": 9.966292134831461e-06,
"loss": 1.5157,
"step": 1890
},
{
"epoch": 2.6,
"learning_rate": 9.910112359550563e-06,
"loss": 1.5836,
"step": 1900
},
{
"epoch": 2.61,
"learning_rate": 9.853932584269664e-06,
"loss": 1.5842,
"step": 1910
},
{
"epoch": 2.62,
"learning_rate": 9.797752808988766e-06,
"loss": 1.4875,
"step": 1920
},
{
"epoch": 2.64,
"learning_rate": 9.741573033707865e-06,
"loss": 1.628,
"step": 1930
},
{
"epoch": 2.65,
"learning_rate": 9.685393258426967e-06,
"loss": 1.5143,
"step": 1940
},
{
"epoch": 2.66,
"learning_rate": 9.629213483146068e-06,
"loss": 1.5946,
"step": 1950
},
{
"epoch": 2.68,
"learning_rate": 9.57303370786517e-06,
"loss": 1.5819,
"step": 1960
},
{
"epoch": 2.69,
"learning_rate": 9.51685393258427e-06,
"loss": 1.4883,
"step": 1970
},
{
"epoch": 2.7,
"learning_rate": 9.460674157303371e-06,
"loss": 1.6313,
"step": 1980
},
{
"epoch": 2.72,
"learning_rate": 9.404494382022474e-06,
"loss": 1.5247,
"step": 1990
},
{
"epoch": 2.73,
"learning_rate": 9.348314606741574e-06,
"loss": 1.5835,
"step": 2000
},
{
"epoch": 2.73,
"eval_loss": 1.6243367195129395,
"eval_runtime": 16.738,
"eval_samples_per_second": 119.488,
"eval_steps_per_second": 1.912,
"step": 2000
},
{
"epoch": 2.75,
"learning_rate": 9.292134831460675e-06,
"loss": 1.5715,
"step": 2010
},
{
"epoch": 2.76,
"learning_rate": 9.235955056179777e-06,
"loss": 1.4771,
"step": 2020
},
{
"epoch": 2.77,
"learning_rate": 9.179775280898876e-06,
"loss": 1.6327,
"step": 2030
},
{
"epoch": 2.79,
"learning_rate": 9.123595505617978e-06,
"loss": 1.5095,
"step": 2040
},
{
"epoch": 2.8,
"learning_rate": 9.06741573033708e-06,
"loss": 1.5819,
"step": 2050
},
{
"epoch": 2.81,
"learning_rate": 9.01123595505618e-06,
"loss": 1.5882,
"step": 2060
},
{
"epoch": 2.83,
"learning_rate": 8.95505617977528e-06,
"loss": 1.4924,
"step": 2070
},
{
"epoch": 2.84,
"learning_rate": 8.898876404494383e-06,
"loss": 1.6345,
"step": 2080
},
{
"epoch": 2.86,
"learning_rate": 8.842696629213485e-06,
"loss": 1.5132,
"step": 2090
},
{
"epoch": 2.87,
"learning_rate": 8.786516853932585e-06,
"loss": 1.5971,
"step": 2100
},
{
"epoch": 2.88,
"learning_rate": 8.730337078651686e-06,
"loss": 1.5864,
"step": 2110
},
{
"epoch": 2.9,
"learning_rate": 8.674157303370788e-06,
"loss": 1.4847,
"step": 2120
},
{
"epoch": 2.91,
"learning_rate": 8.617977528089887e-06,
"loss": 1.6265,
"step": 2130
},
{
"epoch": 2.92,
"learning_rate": 8.561797752808989e-06,
"loss": 1.5086,
"step": 2140
},
{
"epoch": 2.94,
"learning_rate": 8.50561797752809e-06,
"loss": 1.5775,
"step": 2150
},
{
"epoch": 2.95,
"learning_rate": 8.449438202247192e-06,
"loss": 1.5838,
"step": 2160
},
{
"epoch": 2.96,
"learning_rate": 8.393258426966293e-06,
"loss": 1.495,
"step": 2170
},
{
"epoch": 2.98,
"learning_rate": 8.337078651685394e-06,
"loss": 1.6272,
"step": 2180
},
{
"epoch": 2.99,
"learning_rate": 8.280898876404496e-06,
"loss": 1.51,
"step": 2190
},
{
"epoch": 3.01,
"learning_rate": 8.224719101123596e-06,
"loss": 1.5862,
"step": 2200
},
{
"epoch": 3.01,
"eval_loss": 1.62160325050354,
"eval_runtime": 16.7405,
"eval_samples_per_second": 119.471,
"eval_steps_per_second": 1.912,
"step": 2200
},
{
"epoch": 3.02,
"learning_rate": 8.168539325842697e-06,
"loss": 1.552,
"step": 2210
},
{
"epoch": 3.03,
"learning_rate": 8.112359550561798e-06,
"loss": 1.5758,
"step": 2220
},
{
"epoch": 3.05,
"learning_rate": 8.056179775280898e-06,
"loss": 1.5457,
"step": 2230
},
{
"epoch": 3.06,
"learning_rate": 8.000000000000001e-06,
"loss": 1.5222,
"step": 2240
},
{
"epoch": 3.07,
"learning_rate": 7.943820224719103e-06,
"loss": 1.5822,
"step": 2250
},
{
"epoch": 3.09,
"learning_rate": 7.887640449438203e-06,
"loss": 1.5349,
"step": 2260
},
{
"epoch": 3.1,
"learning_rate": 7.831460674157304e-06,
"loss": 1.5731,
"step": 2270
},
{
"epoch": 3.11,
"learning_rate": 7.775280898876405e-06,
"loss": 1.5462,
"step": 2280
},
{
"epoch": 3.13,
"learning_rate": 7.719101123595507e-06,
"loss": 1.5294,
"step": 2290
},
{
"epoch": 3.14,
"learning_rate": 7.662921348314607e-06,
"loss": 1.5766,
"step": 2300
},
{
"epoch": 3.16,
"learning_rate": 7.606741573033708e-06,
"loss": 1.5479,
"step": 2310
},
{
"epoch": 3.17,
"learning_rate": 7.55056179775281e-06,
"loss": 1.5697,
"step": 2320
},
{
"epoch": 3.18,
"learning_rate": 7.494382022471911e-06,
"loss": 1.5292,
"step": 2330
},
{
"epoch": 3.2,
"learning_rate": 7.438202247191012e-06,
"loss": 1.5249,
"step": 2340
},
{
"epoch": 3.21,
"learning_rate": 7.382022471910113e-06,
"loss": 1.5781,
"step": 2350
},
{
"epoch": 3.22,
"learning_rate": 7.325842696629214e-06,
"loss": 1.5538,
"step": 2360
},
{
"epoch": 3.24,
"learning_rate": 7.269662921348315e-06,
"loss": 1.5866,
"step": 2370
},
{
"epoch": 3.25,
"learning_rate": 7.213483146067416e-06,
"loss": 1.5419,
"step": 2380
},
{
"epoch": 3.27,
"learning_rate": 7.157303370786517e-06,
"loss": 1.5232,
"step": 2390
},
{
"epoch": 3.28,
"learning_rate": 7.101123595505619e-06,
"loss": 1.5776,
"step": 2400
},
{
"epoch": 3.28,
"eval_loss": 1.6191555261611938,
"eval_runtime": 16.7368,
"eval_samples_per_second": 119.497,
"eval_steps_per_second": 1.912,
"step": 2400
},
{
"epoch": 3.29,
"learning_rate": 7.04494382022472e-06,
"loss": 1.5493,
"step": 2410
},
{
"epoch": 3.31,
"learning_rate": 6.988764044943821e-06,
"loss": 1.5704,
"step": 2420
},
{
"epoch": 3.32,
"learning_rate": 6.932584269662922e-06,
"loss": 1.5418,
"step": 2430
},
{
"epoch": 3.33,
"learning_rate": 6.876404494382023e-06,
"loss": 1.5322,
"step": 2440
},
{
"epoch": 3.35,
"learning_rate": 6.820224719101124e-06,
"loss": 1.5706,
"step": 2450
},
{
"epoch": 3.36,
"learning_rate": 6.764044943820225e-06,
"loss": 1.552,
"step": 2460
},
{
"epoch": 3.37,
"learning_rate": 6.707865168539326e-06,
"loss": 1.5619,
"step": 2470
},
{
"epoch": 3.39,
"learning_rate": 6.651685393258428e-06,
"loss": 1.546,
"step": 2480
},
{
"epoch": 3.4,
"learning_rate": 6.595505617977529e-06,
"loss": 1.5281,
"step": 2490
},
{
"epoch": 3.42,
"learning_rate": 6.53932584269663e-06,
"loss": 1.5864,
"step": 2500
},
{
"epoch": 3.43,
"learning_rate": 6.483146067415731e-06,
"loss": 1.5541,
"step": 2510
},
{
"epoch": 3.44,
"learning_rate": 6.426966292134832e-06,
"loss": 1.5799,
"step": 2520
},
{
"epoch": 3.46,
"learning_rate": 6.370786516853933e-06,
"loss": 1.5377,
"step": 2530
},
{
"epoch": 3.47,
"learning_rate": 6.314606741573034e-06,
"loss": 1.512,
"step": 2540
},
{
"epoch": 3.48,
"learning_rate": 6.258426966292135e-06,
"loss": 1.58,
"step": 2550
},
{
"epoch": 3.5,
"learning_rate": 6.202247191011237e-06,
"loss": 1.544,
"step": 2560
},
{
"epoch": 3.51,
"learning_rate": 6.146067415730338e-06,
"loss": 1.5729,
"step": 2570
},
{
"epoch": 3.52,
"learning_rate": 6.089887640449439e-06,
"loss": 1.5337,
"step": 2580
},
{
"epoch": 3.54,
"learning_rate": 6.03370786516854e-06,
"loss": 1.5258,
"step": 2590
},
{
"epoch": 3.55,
"learning_rate": 5.977528089887641e-06,
"loss": 1.5803,
"step": 2600
},
{
"epoch": 3.55,
"eval_loss": 1.6172821521759033,
"eval_runtime": 16.7295,
"eval_samples_per_second": 119.549,
"eval_steps_per_second": 1.913,
"step": 2600
},
{
"epoch": 3.57,
"learning_rate": 5.921348314606742e-06,
"loss": 1.5463,
"step": 2610
},
{
"epoch": 3.58,
"learning_rate": 5.865168539325843e-06,
"loss": 1.5667,
"step": 2620
},
{
"epoch": 3.59,
"learning_rate": 5.808988764044944e-06,
"loss": 1.542,
"step": 2630
},
{
"epoch": 3.61,
"learning_rate": 5.752808988764046e-06,
"loss": 1.5207,
"step": 2640
},
{
"epoch": 3.62,
"learning_rate": 5.6966292134831465e-06,
"loss": 1.5787,
"step": 2650
},
{
"epoch": 3.63,
"learning_rate": 5.640449438202248e-06,
"loss": 1.554,
"step": 2660
},
{
"epoch": 3.65,
"learning_rate": 5.5842696629213486e-06,
"loss": 1.5634,
"step": 2670
},
{
"epoch": 3.66,
"learning_rate": 5.52808988764045e-06,
"loss": 1.5399,
"step": 2680
},
{
"epoch": 3.67,
"learning_rate": 5.4719101123595506e-06,
"loss": 1.5271,
"step": 2690
},
{
"epoch": 3.69,
"learning_rate": 5.415730337078652e-06,
"loss": 1.5812,
"step": 2700
},
{
"epoch": 3.7,
"learning_rate": 5.359550561797753e-06,
"loss": 1.5445,
"step": 2710
},
{
"epoch": 3.72,
"learning_rate": 5.303370786516855e-06,
"loss": 1.5694,
"step": 2720
},
{
"epoch": 3.73,
"learning_rate": 5.2471910112359555e-06,
"loss": 1.5422,
"step": 2730
},
{
"epoch": 3.74,
"learning_rate": 5.191011235955057e-06,
"loss": 1.5261,
"step": 2740
},
{
"epoch": 3.76,
"learning_rate": 5.1348314606741575e-06,
"loss": 1.5861,
"step": 2750
},
{
"epoch": 3.77,
"learning_rate": 5.078651685393259e-06,
"loss": 1.5381,
"step": 2760
},
{
"epoch": 3.78,
"learning_rate": 5.0224719101123595e-06,
"loss": 1.5697,
"step": 2770
},
{
"epoch": 3.8,
"learning_rate": 4.966292134831461e-06,
"loss": 1.5373,
"step": 2780
},
{
"epoch": 3.81,
"learning_rate": 4.910112359550562e-06,
"loss": 1.5225,
"step": 2790
},
{
"epoch": 3.83,
"learning_rate": 4.853932584269663e-06,
"loss": 1.5747,
"step": 2800
},
{
"epoch": 3.83,
"eval_loss": 1.6157482862472534,
"eval_runtime": 16.7399,
"eval_samples_per_second": 119.475,
"eval_steps_per_second": 1.912,
"step": 2800
},
{
"epoch": 3.84,
"learning_rate": 4.797752808988764e-06,
"loss": 1.5451,
"step": 2810
},
{
"epoch": 3.85,
"learning_rate": 4.741573033707866e-06,
"loss": 1.5685,
"step": 2820
},
{
"epoch": 3.87,
"learning_rate": 4.685393258426966e-06,
"loss": 1.5349,
"step": 2830
},
{
"epoch": 3.88,
"learning_rate": 4.629213483146068e-06,
"loss": 1.5274,
"step": 2840
},
{
"epoch": 3.89,
"learning_rate": 4.573033707865168e-06,
"loss": 1.5779,
"step": 2850
},
{
"epoch": 3.91,
"learning_rate": 4.516853932584271e-06,
"loss": 1.546,
"step": 2860
},
{
"epoch": 3.92,
"learning_rate": 4.460674157303371e-06,
"loss": 1.57,
"step": 2870
},
{
"epoch": 3.93,
"learning_rate": 4.404494382022472e-06,
"loss": 1.5292,
"step": 2880
},
{
"epoch": 3.95,
"learning_rate": 4.348314606741573e-06,
"loss": 1.5177,
"step": 2890
},
{
"epoch": 3.96,
"learning_rate": 4.292134831460675e-06,
"loss": 1.5847,
"step": 2900
},
{
"epoch": 3.98,
"learning_rate": 4.235955056179776e-06,
"loss": 1.5438,
"step": 2910
},
{
"epoch": 3.99,
"learning_rate": 4.179775280898877e-06,
"loss": 1.563,
"step": 2920
},
{
"epoch": 4.0,
"learning_rate": 4.123595505617977e-06,
"loss": 1.5172,
"step": 2930
},
{
"epoch": 4.02,
"learning_rate": 4.06741573033708e-06,
"loss": 1.5166,
"step": 2940
},
{
"epoch": 4.03,
"learning_rate": 4.01123595505618e-06,
"loss": 1.6083,
"step": 2950
},
{
"epoch": 4.04,
"learning_rate": 3.955056179775282e-06,
"loss": 1.4814,
"step": 2960
},
{
"epoch": 4.06,
"learning_rate": 3.898876404494382e-06,
"loss": 1.5859,
"step": 2970
},
{
"epoch": 4.07,
"learning_rate": 3.842696629213483e-06,
"loss": 1.5363,
"step": 2980
},
{
"epoch": 4.08,
"learning_rate": 3.7865168539325847e-06,
"loss": 1.5146,
"step": 2990
},
{
"epoch": 4.1,
"learning_rate": 3.7303370786516857e-06,
"loss": 1.6161,
"step": 3000
},
{
"epoch": 4.1,
"eval_loss": 1.6143279075622559,
"eval_runtime": 16.715,
"eval_samples_per_second": 119.653,
"eval_steps_per_second": 1.914,
"step": 3000
},
{
"epoch": 4.11,
"learning_rate": 3.6741573033707867e-06,
"loss": 1.479,
"step": 3010
},
{
"epoch": 4.13,
"learning_rate": 3.6179775280898877e-06,
"loss": 1.5963,
"step": 3020
},
{
"epoch": 4.14,
"learning_rate": 3.561797752808989e-06,
"loss": 1.5339,
"step": 3030
},
{
"epoch": 4.15,
"learning_rate": 3.50561797752809e-06,
"loss": 1.5162,
"step": 3040
},
{
"epoch": 4.17,
"learning_rate": 3.449438202247191e-06,
"loss": 1.6136,
"step": 3050
},
{
"epoch": 4.18,
"learning_rate": 3.393258426966292e-06,
"loss": 1.4947,
"step": 3060
},
{
"epoch": 4.19,
"learning_rate": 3.3370786516853936e-06,
"loss": 1.5975,
"step": 3070
},
{
"epoch": 4.21,
"learning_rate": 3.2808988764044946e-06,
"loss": 1.5325,
"step": 3080
},
{
"epoch": 4.22,
"learning_rate": 3.2247191011235956e-06,
"loss": 1.5184,
"step": 3090
},
{
"epoch": 4.23,
"learning_rate": 3.1685393258426966e-06,
"loss": 1.6166,
"step": 3100
},
{
"epoch": 4.25,
"learning_rate": 3.112359550561798e-06,
"loss": 1.4882,
"step": 3110
},
{
"epoch": 4.26,
"learning_rate": 3.056179775280899e-06,
"loss": 1.6046,
"step": 3120
},
{
"epoch": 4.28,
"learning_rate": 3e-06,
"loss": 1.5328,
"step": 3130
},
{
"epoch": 4.29,
"learning_rate": 2.943820224719101e-06,
"loss": 1.5216,
"step": 3140
},
{
"epoch": 4.3,
"learning_rate": 2.8876404494382025e-06,
"loss": 1.6175,
"step": 3150
},
{
"epoch": 4.32,
"learning_rate": 2.8314606741573035e-06,
"loss": 1.4852,
"step": 3160
},
{
"epoch": 4.33,
"learning_rate": 2.7752808988764045e-06,
"loss": 1.5937,
"step": 3170
},
{
"epoch": 4.34,
"learning_rate": 2.7191011235955055e-06,
"loss": 1.5304,
"step": 3180
},
{
"epoch": 4.36,
"learning_rate": 2.662921348314607e-06,
"loss": 1.5136,
"step": 3190
},
{
"epoch": 4.37,
"learning_rate": 2.606741573033708e-06,
"loss": 1.6178,
"step": 3200
},
{
"epoch": 4.37,
"eval_loss": 1.613547444343567,
"eval_runtime": 16.7348,
"eval_samples_per_second": 119.512,
"eval_steps_per_second": 1.912,
"step": 3200
},
{
"epoch": 4.39,
"learning_rate": 2.550561797752809e-06,
"loss": 1.4869,
"step": 3210
},
{
"epoch": 4.4,
"learning_rate": 2.4943820224719104e-06,
"loss": 1.6043,
"step": 3220
},
{
"epoch": 4.41,
"learning_rate": 2.4382022471910114e-06,
"loss": 1.5329,
"step": 3230
},
{
"epoch": 4.43,
"learning_rate": 2.3820224719101125e-06,
"loss": 1.5198,
"step": 3240
},
{
"epoch": 4.44,
"learning_rate": 2.3258426966292135e-06,
"loss": 1.6121,
"step": 3250
},
{
"epoch": 4.45,
"learning_rate": 2.269662921348315e-06,
"loss": 1.4784,
"step": 3260
},
{
"epoch": 4.47,
"learning_rate": 2.213483146067416e-06,
"loss": 1.5905,
"step": 3270
},
{
"epoch": 4.48,
"learning_rate": 2.157303370786517e-06,
"loss": 1.525,
"step": 3280
},
{
"epoch": 4.49,
"learning_rate": 2.101123595505618e-06,
"loss": 1.5186,
"step": 3290
},
{
"epoch": 4.51,
"learning_rate": 2.0449438202247194e-06,
"loss": 1.6125,
"step": 3300
},
{
"epoch": 4.52,
"learning_rate": 1.9887640449438204e-06,
"loss": 1.4823,
"step": 3310
},
{
"epoch": 4.54,
"learning_rate": 1.9325842696629214e-06,
"loss": 1.5908,
"step": 3320
},
{
"epoch": 4.55,
"learning_rate": 1.8764044943820226e-06,
"loss": 1.5302,
"step": 3330
},
{
"epoch": 4.56,
"learning_rate": 1.8202247191011238e-06,
"loss": 1.5221,
"step": 3340
},
{
"epoch": 4.58,
"learning_rate": 1.7640449438202248e-06,
"loss": 1.6119,
"step": 3350
},
{
"epoch": 4.59,
"learning_rate": 1.707865168539326e-06,
"loss": 1.4835,
"step": 3360
},
{
"epoch": 4.6,
"learning_rate": 1.651685393258427e-06,
"loss": 1.5988,
"step": 3370
},
{
"epoch": 4.62,
"learning_rate": 1.5955056179775283e-06,
"loss": 1.5275,
"step": 3380
},
{
"epoch": 4.63,
"learning_rate": 1.5393258426966293e-06,
"loss": 1.5237,
"step": 3390
},
{
"epoch": 4.64,
"learning_rate": 1.4831460674157305e-06,
"loss": 1.6187,
"step": 3400
},
{
"epoch": 4.64,
"eval_loss": 1.6129449605941772,
"eval_runtime": 16.7334,
"eval_samples_per_second": 119.521,
"eval_steps_per_second": 1.912,
"step": 3400
}
],
"max_steps": 3660,
"num_train_epochs": 5,
"total_flos": 2.4426149740214747e+19,
"trial_name": null,
"trial_params": null
}