lora-copa / trainer_state.json
igzi's picture
Upload latest checkpoint for copa
37f74c6 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 7.142857142857143,
"eval_steps": 500,
"global_step": 5000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.07142857142857142,
"grad_norm": 0.0002574920654296875,
"learning_rate": 0.00019800000000000002,
"loss": 0.7765,
"step": 50
},
{
"epoch": 0.14285714285714285,
"grad_norm": 0.00133514404296875,
"learning_rate": 0.000196,
"loss": 0.3218,
"step": 100
},
{
"epoch": 0.21428571428571427,
"grad_norm": 0.01507568359375,
"learning_rate": 0.000194,
"loss": 0.3479,
"step": 150
},
{
"epoch": 0.2857142857142857,
"grad_norm": 0.000885009765625,
"learning_rate": 0.000192,
"loss": 0.5917,
"step": 200
},
{
"epoch": 0.35714285714285715,
"grad_norm": 4.65625,
"learning_rate": 0.00019,
"loss": 0.5147,
"step": 250
},
{
"epoch": 0.42857142857142855,
"grad_norm": 0.002349853515625,
"learning_rate": 0.000188,
"loss": 0.2799,
"step": 300
},
{
"epoch": 0.5,
"grad_norm": 8.7738037109375e-05,
"learning_rate": 0.00018600000000000002,
"loss": 0.3272,
"step": 350
},
{
"epoch": 0.5714285714285714,
"grad_norm": 0.000308990478515625,
"learning_rate": 0.00018400000000000003,
"loss": 0.4057,
"step": 400
},
{
"epoch": 0.6428571428571429,
"grad_norm": 3.734375,
"learning_rate": 0.000182,
"loss": 0.665,
"step": 450
},
{
"epoch": 0.7142857142857143,
"grad_norm": 0.212890625,
"learning_rate": 0.00018,
"loss": 0.3702,
"step": 500
},
{
"epoch": 0.7857142857142857,
"grad_norm": 0.0162353515625,
"learning_rate": 0.00017800000000000002,
"loss": 0.5988,
"step": 550
},
{
"epoch": 0.8571428571428571,
"grad_norm": 6.961822509765625e-05,
"learning_rate": 0.00017600000000000002,
"loss": 0.1097,
"step": 600
},
{
"epoch": 0.9285714285714286,
"grad_norm": 8.7738037109375e-05,
"learning_rate": 0.000174,
"loss": 0.362,
"step": 650
},
{
"epoch": 1.0,
"grad_norm": 2.6941299438476562e-05,
"learning_rate": 0.000172,
"loss": 0.2517,
"step": 700
},
{
"epoch": 1.0714285714285714,
"grad_norm": 0.000732421875,
"learning_rate": 0.00017,
"loss": 0.1627,
"step": 750
},
{
"epoch": 1.1428571428571428,
"grad_norm": 0.0002727508544921875,
"learning_rate": 0.000168,
"loss": 0.1745,
"step": 800
},
{
"epoch": 1.2142857142857142,
"grad_norm": 0.0002079010009765625,
"learning_rate": 0.000166,
"loss": 0.2209,
"step": 850
},
{
"epoch": 1.2857142857142856,
"grad_norm": 0.0003185272216796875,
"learning_rate": 0.000164,
"loss": 0.1962,
"step": 900
},
{
"epoch": 1.3571428571428572,
"grad_norm": 0.0001583099365234375,
"learning_rate": 0.000162,
"loss": 0.221,
"step": 950
},
{
"epoch": 1.4285714285714286,
"grad_norm": 1.9375,
"learning_rate": 0.00016,
"loss": 0.2535,
"step": 1000
},
{
"epoch": 1.5,
"grad_norm": 0.00011110305786132812,
"learning_rate": 0.00015800000000000002,
"loss": 0.2031,
"step": 1050
},
{
"epoch": 1.5714285714285714,
"grad_norm": 0.00010919570922851562,
"learning_rate": 0.00015600000000000002,
"loss": 0.1939,
"step": 1100
},
{
"epoch": 1.6428571428571428,
"grad_norm": 6.961822509765625e-05,
"learning_rate": 0.000154,
"loss": 0.2314,
"step": 1150
},
{
"epoch": 1.7142857142857144,
"grad_norm": 8.296966552734375e-05,
"learning_rate": 0.000152,
"loss": 0.1713,
"step": 1200
},
{
"epoch": 1.7857142857142856,
"grad_norm": 5.054473876953125e-05,
"learning_rate": 0.00015000000000000001,
"loss": 0.1591,
"step": 1250
},
{
"epoch": 1.8571428571428572,
"grad_norm": 4.375,
"learning_rate": 0.000148,
"loss": 0.162,
"step": 1300
},
{
"epoch": 1.9285714285714286,
"grad_norm": 0.0013427734375,
"learning_rate": 0.000146,
"loss": 0.1312,
"step": 1350
},
{
"epoch": 2.0,
"grad_norm": 4.029273986816406e-05,
"learning_rate": 0.000144,
"loss": 0.2479,
"step": 1400
},
{
"epoch": 2.0714285714285716,
"grad_norm": 0.00019073486328125,
"learning_rate": 0.000142,
"loss": 0.1328,
"step": 1450
},
{
"epoch": 2.142857142857143,
"grad_norm": 5.030632019042969e-05,
"learning_rate": 0.00014,
"loss": 0.0929,
"step": 1500
},
{
"epoch": 2.2142857142857144,
"grad_norm": 1.53125,
"learning_rate": 0.000138,
"loss": 0.1952,
"step": 1550
},
{
"epoch": 2.2857142857142856,
"grad_norm": 9.918212890625e-05,
"learning_rate": 0.00013600000000000003,
"loss": 0.1879,
"step": 1600
},
{
"epoch": 2.357142857142857,
"grad_norm": 5.46875,
"learning_rate": 0.000134,
"loss": 0.1255,
"step": 1650
},
{
"epoch": 2.4285714285714284,
"grad_norm": 0.00020694732666015625,
"learning_rate": 0.000132,
"loss": 0.1906,
"step": 1700
},
{
"epoch": 2.5,
"grad_norm": 0.0035858154296875,
"learning_rate": 0.00013000000000000002,
"loss": 0.0698,
"step": 1750
},
{
"epoch": 2.571428571428571,
"grad_norm": 8.0108642578125e-05,
"learning_rate": 0.00012800000000000002,
"loss": 0.1395,
"step": 1800
},
{
"epoch": 2.642857142857143,
"grad_norm": 0.000118255615234375,
"learning_rate": 0.000126,
"loss": 0.1277,
"step": 1850
},
{
"epoch": 2.7142857142857144,
"grad_norm": 0.000347137451171875,
"learning_rate": 0.000124,
"loss": 0.0916,
"step": 1900
},
{
"epoch": 2.7857142857142856,
"grad_norm": 4.458427429199219e-05,
"learning_rate": 0.000122,
"loss": 0.1283,
"step": 1950
},
{
"epoch": 2.857142857142857,
"grad_norm": 0.000820159912109375,
"learning_rate": 0.00012,
"loss": 0.097,
"step": 2000
},
{
"epoch": 2.928571428571429,
"grad_norm": 4.601478576660156e-05,
"learning_rate": 0.000118,
"loss": 0.0421,
"step": 2050
},
{
"epoch": 3.0,
"grad_norm": 7.09375,
"learning_rate": 0.000116,
"loss": 0.122,
"step": 2100
},
{
"epoch": 3.0714285714285716,
"grad_norm": 1.90625,
"learning_rate": 0.00011399999999999999,
"loss": 0.1375,
"step": 2150
},
{
"epoch": 3.142857142857143,
"grad_norm": 9.72747802734375e-05,
"learning_rate": 0.00011200000000000001,
"loss": 0.1241,
"step": 2200
},
{
"epoch": 3.2142857142857144,
"grad_norm": 0.0001316070556640625,
"learning_rate": 0.00011000000000000002,
"loss": 0.125,
"step": 2250
},
{
"epoch": 3.2857142857142856,
"grad_norm": 0.00015735626220703125,
"learning_rate": 0.00010800000000000001,
"loss": 0.0277,
"step": 2300
},
{
"epoch": 3.357142857142857,
"grad_norm": 0.00057220458984375,
"learning_rate": 0.00010600000000000002,
"loss": 0.0974,
"step": 2350
},
{
"epoch": 3.4285714285714284,
"grad_norm": 0.00099945068359375,
"learning_rate": 0.00010400000000000001,
"loss": 0.083,
"step": 2400
},
{
"epoch": 3.5,
"grad_norm": 0.000179290771484375,
"learning_rate": 0.00010200000000000001,
"loss": 0.0856,
"step": 2450
},
{
"epoch": 3.571428571428571,
"grad_norm": 0.00011014938354492188,
"learning_rate": 0.0001,
"loss": 0.1169,
"step": 2500
},
{
"epoch": 3.642857142857143,
"grad_norm": 1.625,
"learning_rate": 9.8e-05,
"loss": 0.061,
"step": 2550
},
{
"epoch": 3.7142857142857144,
"grad_norm": 0.000579833984375,
"learning_rate": 9.6e-05,
"loss": 0.1022,
"step": 2600
},
{
"epoch": 3.7857142857142856,
"grad_norm": 2.1696090698242188e-05,
"learning_rate": 9.4e-05,
"loss": 0.0985,
"step": 2650
},
{
"epoch": 3.857142857142857,
"grad_norm": 0.00012969970703125,
"learning_rate": 9.200000000000001e-05,
"loss": 0.0822,
"step": 2700
},
{
"epoch": 3.928571428571429,
"grad_norm": 1.7762184143066406e-05,
"learning_rate": 9e-05,
"loss": 0.1027,
"step": 2750
},
{
"epoch": 4.0,
"grad_norm": 0.0002956390380859375,
"learning_rate": 8.800000000000001e-05,
"loss": 0.0908,
"step": 2800
},
{
"epoch": 4.071428571428571,
"grad_norm": 5.125,
"learning_rate": 8.6e-05,
"loss": 0.055,
"step": 2850
},
{
"epoch": 4.142857142857143,
"grad_norm": 2.234375,
"learning_rate": 8.4e-05,
"loss": 0.0411,
"step": 2900
},
{
"epoch": 4.214285714285714,
"grad_norm": 1.704692840576172e-05,
"learning_rate": 8.2e-05,
"loss": 0.0862,
"step": 2950
},
{
"epoch": 4.285714285714286,
"grad_norm": 0.0030517578125,
"learning_rate": 8e-05,
"loss": 0.0639,
"step": 3000
},
{
"epoch": 4.357142857142857,
"grad_norm": 1.7642974853515625e-05,
"learning_rate": 7.800000000000001e-05,
"loss": 0.0855,
"step": 3050
},
{
"epoch": 4.428571428571429,
"grad_norm": 6.818771362304688e-05,
"learning_rate": 7.6e-05,
"loss": 0.0793,
"step": 3100
},
{
"epoch": 4.5,
"grad_norm": 4.0,
"learning_rate": 7.4e-05,
"loss": 0.1272,
"step": 3150
},
{
"epoch": 4.571428571428571,
"grad_norm": 3.382563591003418e-06,
"learning_rate": 7.2e-05,
"loss": 0.0904,
"step": 3200
},
{
"epoch": 4.642857142857143,
"grad_norm": 2.140625,
"learning_rate": 7e-05,
"loss": 0.0605,
"step": 3250
},
{
"epoch": 4.714285714285714,
"grad_norm": 3.0040740966796875e-05,
"learning_rate": 6.800000000000001e-05,
"loss": 0.1199,
"step": 3300
},
{
"epoch": 4.785714285714286,
"grad_norm": 0.00037384033203125,
"learning_rate": 6.6e-05,
"loss": 0.1049,
"step": 3350
},
{
"epoch": 4.857142857142857,
"grad_norm": 9.918212890625e-05,
"learning_rate": 6.400000000000001e-05,
"loss": 0.0821,
"step": 3400
},
{
"epoch": 4.928571428571429,
"grad_norm": 0.00156402587890625,
"learning_rate": 6.2e-05,
"loss": 0.0807,
"step": 3450
},
{
"epoch": 5.0,
"grad_norm": 0.001068115234375,
"learning_rate": 6e-05,
"loss": 0.084,
"step": 3500
},
{
"epoch": 5.071428571428571,
"grad_norm": 4.124641418457031e-05,
"learning_rate": 5.8e-05,
"loss": 0.0922,
"step": 3550
},
{
"epoch": 5.142857142857143,
"grad_norm": 1.5854835510253906e-05,
"learning_rate": 5.6000000000000006e-05,
"loss": 0.058,
"step": 3600
},
{
"epoch": 5.214285714285714,
"grad_norm": 4.03125,
"learning_rate": 5.4000000000000005e-05,
"loss": 0.0838,
"step": 3650
},
{
"epoch": 5.285714285714286,
"grad_norm": 4.28125,
"learning_rate": 5.2000000000000004e-05,
"loss": 0.0592,
"step": 3700
},
{
"epoch": 5.357142857142857,
"grad_norm": 1.3172626495361328e-05,
"learning_rate": 5e-05,
"loss": 0.0694,
"step": 3750
},
{
"epoch": 5.428571428571429,
"grad_norm": 0.00013637542724609375,
"learning_rate": 4.8e-05,
"loss": 0.0848,
"step": 3800
},
{
"epoch": 5.5,
"grad_norm": 0.00014495849609375,
"learning_rate": 4.600000000000001e-05,
"loss": 0.1067,
"step": 3850
},
{
"epoch": 5.571428571428571,
"grad_norm": 0.0006256103515625,
"learning_rate": 4.4000000000000006e-05,
"loss": 0.0789,
"step": 3900
},
{
"epoch": 5.642857142857143,
"grad_norm": 2.171875,
"learning_rate": 4.2e-05,
"loss": 0.0608,
"step": 3950
},
{
"epoch": 5.714285714285714,
"grad_norm": 0.000453948974609375,
"learning_rate": 4e-05,
"loss": 0.0327,
"step": 4000
},
{
"epoch": 5.785714285714286,
"grad_norm": 0.0003566741943359375,
"learning_rate": 3.8e-05,
"loss": 0.085,
"step": 4050
},
{
"epoch": 5.857142857142857,
"grad_norm": 2.47955322265625e-05,
"learning_rate": 3.6e-05,
"loss": 0.1033,
"step": 4100
},
{
"epoch": 5.928571428571429,
"grad_norm": 5.221366882324219e-05,
"learning_rate": 3.4000000000000007e-05,
"loss": 0.0794,
"step": 4150
},
{
"epoch": 6.0,
"grad_norm": 8.392333984375e-05,
"learning_rate": 3.2000000000000005e-05,
"loss": 0.0581,
"step": 4200
},
{
"epoch": 6.071428571428571,
"grad_norm": 0.0001659393310546875,
"learning_rate": 3e-05,
"loss": 0.033,
"step": 4250
},
{
"epoch": 6.142857142857143,
"grad_norm": 0.0004634857177734375,
"learning_rate": 2.8000000000000003e-05,
"loss": 0.0904,
"step": 4300
},
{
"epoch": 6.214285714285714,
"grad_norm": 0.00020503997802734375,
"learning_rate": 2.6000000000000002e-05,
"loss": 0.086,
"step": 4350
},
{
"epoch": 6.285714285714286,
"grad_norm": 6.079673767089844e-05,
"learning_rate": 2.4e-05,
"loss": 0.0653,
"step": 4400
},
{
"epoch": 6.357142857142857,
"grad_norm": 8.440017700195312e-05,
"learning_rate": 2.2000000000000003e-05,
"loss": 0.0519,
"step": 4450
},
{
"epoch": 6.428571428571429,
"grad_norm": 1.2578125,
"learning_rate": 2e-05,
"loss": 0.0781,
"step": 4500
},
{
"epoch": 6.5,
"grad_norm": 0.00049591064453125,
"learning_rate": 1.8e-05,
"loss": 0.0625,
"step": 4550
},
{
"epoch": 6.571428571428571,
"grad_norm": 8.106231689453125e-05,
"learning_rate": 1.6000000000000003e-05,
"loss": 0.0884,
"step": 4600
},
{
"epoch": 6.642857142857143,
"grad_norm": 2.8967857360839844e-05,
"learning_rate": 1.4000000000000001e-05,
"loss": 0.0893,
"step": 4650
},
{
"epoch": 6.714285714285714,
"grad_norm": 1.055002212524414e-05,
"learning_rate": 1.2e-05,
"loss": 0.0631,
"step": 4700
},
{
"epoch": 6.785714285714286,
"grad_norm": 7.009506225585938e-05,
"learning_rate": 1e-05,
"loss": 0.0841,
"step": 4750
},
{
"epoch": 6.857142857142857,
"grad_norm": 0.0003681182861328125,
"learning_rate": 8.000000000000001e-06,
"loss": 0.0794,
"step": 4800
},
{
"epoch": 6.928571428571429,
"grad_norm": 0.00031280517578125,
"learning_rate": 6e-06,
"loss": 0.0826,
"step": 4850
},
{
"epoch": 7.0,
"grad_norm": 2.944469451904297e-05,
"learning_rate": 4.000000000000001e-06,
"loss": 0.0306,
"step": 4900
},
{
"epoch": 7.071428571428571,
"grad_norm": 0.000453948974609375,
"learning_rate": 2.0000000000000003e-06,
"loss": 0.0654,
"step": 4950
},
{
"epoch": 7.142857142857143,
"grad_norm": 0.0001373291015625,
"learning_rate": 0.0,
"loss": 0.0638,
"step": 5000
}
],
"logging_steps": 50,
"max_steps": 5000,
"num_input_tokens_seen": 0,
"num_train_epochs": 8,
"save_steps": 1000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 1.7611786549248e+16,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}