arpandeepk's picture
upload rmu_Llama-3.1-8B_20260427_203908
77e8fad verified
Raw
History Blame
16.6 kB
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.7511737089201878,
"eval_steps": 500,
"global_step": 200,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0,
"forget_loss": 0.00946044921875,
"retain_loss": 0.0,
"step": 0
},
{
"epoch": 0,
"forget_loss": 0.00994873046875,
"retain_loss": 0.0,
"step": 0
},
{
"epoch": 0,
"forget_loss": 0.00958251953125,
"retain_loss": 0.0,
"step": 0
},
{
"epoch": 0,
"forget_loss": 0.009521484375,
"retain_loss": 0.0,
"step": 0
},
{
"epoch": 0.003755868544600939,
"grad_norm": 0.11474609375,
"learning_rate": 0.0,
"loss": 0.0096282958984375,
"step": 1
},
{
"epoch": 0.03755868544600939,
"grad_norm": 0.09814453125,
"learning_rate": 4.5e-06,
"loss": 0.009236653645833334,
"step": 10
},
{
"epoch": 0.03755868544600939,
"forget_loss": 0.00799560546875,
"retain_loss": 1.919269561767578e-05,
"step": 10
},
{
"epoch": 0.03755868544600939,
"forget_loss": 0.008056640625,
"retain_loss": 1.6570091247558594e-05,
"step": 10
},
{
"epoch": 0.03755868544600939,
"forget_loss": 0.0081787109375,
"retain_loss": 2.2649765014648438e-05,
"step": 10
},
{
"epoch": 0.03755868544600939,
"forget_loss": 0.00823974609375,
"retain_loss": 2.1457672119140625e-05,
"step": 10
},
{
"epoch": 0.07511737089201878,
"grad_norm": 0.047607421875,
"learning_rate": 9.5e-06,
"loss": 0.00635223388671875,
"step": 20
},
{
"epoch": 0.07511737089201878,
"forget_loss": 0.003662109375,
"retain_loss": 0.00022602081298828125,
"step": 20
},
{
"epoch": 0.07511737089201878,
"forget_loss": 0.004425048828125,
"retain_loss": 0.0002880096435546875,
"step": 20
},
{
"epoch": 0.07511737089201878,
"forget_loss": 0.0037078857421875,
"retain_loss": 0.00020885467529296875,
"step": 20
},
{
"epoch": 0.07511737089201878,
"forget_loss": 0.00421142578125,
"retain_loss": 0.00021457672119140625,
"step": 20
},
{
"epoch": 0.11267605633802817,
"grad_norm": 0.123046875,
"learning_rate": 9.82490272373541e-06,
"loss": 0.00320587158203125,
"step": 30
},
{
"epoch": 0.11267605633802817,
"forget_loss": 0.0018310546875,
"retain_loss": 0.00054168701171875,
"step": 30
},
{
"epoch": 0.11267605633802817,
"forget_loss": 0.00177001953125,
"retain_loss": 0.0004177093505859375,
"step": 30
},
{
"epoch": 0.11267605633802817,
"forget_loss": 0.00177001953125,
"retain_loss": 0.0002956390380859375,
"step": 30
},
{
"epoch": 0.11267605633802817,
"forget_loss": 0.0016937255859375,
"retain_loss": 0.000331878662109375,
"step": 30
},
{
"epoch": 0.15023474178403756,
"grad_norm": 0.05517578125,
"learning_rate": 9.63035019455253e-06,
"loss": 0.0016145706176757812,
"step": 40
},
{
"epoch": 0.15023474178403756,
"forget_loss": 0.0007781982421875,
"retain_loss": 0.00021648406982421875,
"step": 40
},
{
"epoch": 0.15023474178403756,
"forget_loss": 0.000762939453125,
"retain_loss": 0.0004100799560546875,
"step": 40
},
{
"epoch": 0.15023474178403756,
"forget_loss": 0.000766754150390625,
"retain_loss": 0.000331878662109375,
"step": 40
},
{
"epoch": 0.15023474178403756,
"forget_loss": 0.0007781982421875,
"retain_loss": 0.00038909912109375,
"step": 40
},
{
"epoch": 0.18779342723004694,
"grad_norm": 0.1416015625,
"learning_rate": 9.435797665369651e-06,
"loss": 0.000954437255859375,
"step": 50
},
{
"epoch": 0.18779342723004694,
"forget_loss": 0.0003452301025390625,
"retain_loss": 0.00020599365234375,
"step": 50
},
{
"epoch": 0.18779342723004694,
"forget_loss": 0.0003814697265625,
"retain_loss": 0.00020694732666015625,
"step": 50
},
{
"epoch": 0.18779342723004694,
"forget_loss": 0.000415802001953125,
"retain_loss": 0.0003604888916015625,
"step": 50
},
{
"epoch": 0.18779342723004694,
"forget_loss": 0.0003719329833984375,
"retain_loss": 0.0001964569091796875,
"step": 50
},
{
"epoch": 0.22535211267605634,
"grad_norm": 0.06396484375,
"learning_rate": 9.241245136186772e-06,
"loss": 0.0008329868316650391,
"step": 60
},
{
"epoch": 0.22535211267605634,
"forget_loss": 0.0002574920654296875,
"retain_loss": 0.0002727508544921875,
"step": 60
},
{
"epoch": 0.22535211267605634,
"forget_loss": 0.00019931793212890625,
"retain_loss": 0.00010919570922851562,
"step": 60
},
{
"epoch": 0.22535211267605634,
"forget_loss": 0.0001983642578125,
"retain_loss": 0.000202178955078125,
"step": 60
},
{
"epoch": 0.22535211267605634,
"forget_loss": 0.00023365020751953125,
"retain_loss": 0.0002002716064453125,
"step": 60
},
{
"epoch": 0.26291079812206575,
"grad_norm": 0.28515625,
"learning_rate": 9.046692607003892e-06,
"loss": 0.0006081581115722657,
"step": 70
},
{
"epoch": 0.26291079812206575,
"forget_loss": 0.00018405914306640625,
"retain_loss": 0.0002689361572265625,
"step": 70
},
{
"epoch": 0.26291079812206575,
"forget_loss": 0.0001811981201171875,
"retain_loss": 8.535385131835938e-05,
"step": 70
},
{
"epoch": 0.26291079812206575,
"forget_loss": 0.00022792816162109375,
"retain_loss": 0.00010728836059570312,
"step": 70
},
{
"epoch": 0.26291079812206575,
"forget_loss": 0.0003643035888671875,
"retain_loss": 0.00011777877807617188,
"step": 70
},
{
"epoch": 0.3004694835680751,
"grad_norm": 0.07177734375,
"learning_rate": 8.852140077821012e-06,
"loss": 0.0004550933837890625,
"step": 80
},
{
"epoch": 0.3004694835680751,
"forget_loss": 0.0002155303955078125,
"retain_loss": 0.000110626220703125,
"step": 80
},
{
"epoch": 0.3004694835680751,
"forget_loss": 0.00014019012451171875,
"retain_loss": 6.723403930664062e-05,
"step": 80
},
{
"epoch": 0.3004694835680751,
"forget_loss": 0.0001773834228515625,
"retain_loss": 0.000331878662109375,
"step": 80
},
{
"epoch": 0.3004694835680751,
"forget_loss": 0.00016498565673828125,
"retain_loss": 9.489059448242188e-05,
"step": 80
},
{
"epoch": 0.3380281690140845,
"grad_norm": 0.0810546875,
"learning_rate": 8.657587548638134e-06,
"loss": 0.0003980875015258789,
"step": 90
},
{
"epoch": 0.3380281690140845,
"forget_loss": 0.00015735626220703125,
"retain_loss": 0.0001544952392578125,
"step": 90
},
{
"epoch": 0.3380281690140845,
"forget_loss": 0.000133514404296875,
"retain_loss": 0.00080108642578125,
"step": 90
},
{
"epoch": 0.3380281690140845,
"forget_loss": 0.00014781951904296875,
"retain_loss": 0.000431060791015625,
"step": 90
},
{
"epoch": 0.3380281690140845,
"forget_loss": 0.00014400482177734375,
"retain_loss": 0.00031280517578125,
"step": 90
},
{
"epoch": 0.3755868544600939,
"grad_norm": 0.025146484375,
"learning_rate": 8.463035019455253e-06,
"loss": 0.00032124519348144533,
"step": 100
},
{
"epoch": 0.3755868544600939,
"forget_loss": 0.00015163421630859375,
"retain_loss": 0.00015926361083984375,
"step": 100
},
{
"epoch": 0.3755868544600939,
"forget_loss": 0.0001506805419921875,
"retain_loss": 0.00011587142944335938,
"step": 100
},
{
"epoch": 0.3755868544600939,
"forget_loss": 0.00013065338134765625,
"retain_loss": 0.000125885009765625,
"step": 100
},
{
"epoch": 0.3755868544600939,
"forget_loss": 0.0001239776611328125,
"retain_loss": 0.0001277923583984375,
"step": 100
},
{
"epoch": 0.4131455399061033,
"grad_norm": 0.06884765625,
"learning_rate": 8.268482490272373e-06,
"loss": 0.00034072399139404295,
"step": 110
},
{
"epoch": 0.4131455399061033,
"forget_loss": 0.0001544952392578125,
"retain_loss": 7.2479248046875e-05,
"step": 110
},
{
"epoch": 0.4131455399061033,
"forget_loss": 0.0001373291015625,
"retain_loss": 0.000270843505859375,
"step": 110
},
{
"epoch": 0.4131455399061033,
"forget_loss": 0.0001125335693359375,
"retain_loss": 9.965896606445312e-05,
"step": 110
},
{
"epoch": 0.4131455399061033,
"forget_loss": 0.00011205673217773438,
"retain_loss": 6.866455078125e-05,
"step": 110
},
{
"epoch": 0.4507042253521127,
"grad_norm": 0.08544921875,
"learning_rate": 8.073929961089495e-06,
"loss": 0.0002372264862060547,
"step": 120
},
{
"epoch": 0.4507042253521127,
"forget_loss": 0.000179290771484375,
"retain_loss": 7.05718994140625e-05,
"step": 120
},
{
"epoch": 0.4507042253521127,
"forget_loss": 0.0001277923583984375,
"retain_loss": 9.250640869140625e-05,
"step": 120
},
{
"epoch": 0.4507042253521127,
"forget_loss": 0.00013446807861328125,
"retain_loss": 8.630752563476562e-05,
"step": 120
},
{
"epoch": 0.4507042253521127,
"forget_loss": 0.0001735687255859375,
"retain_loss": 8.106231689453125e-05,
"step": 120
},
{
"epoch": 0.48826291079812206,
"grad_norm": 0.0361328125,
"learning_rate": 7.879377431906614e-06,
"loss": 0.0002105236053466797,
"step": 130
},
{
"epoch": 0.48826291079812206,
"forget_loss": 0.00010776519775390625,
"retain_loss": 5.221366882324219e-05,
"step": 130
},
{
"epoch": 0.48826291079812206,
"forget_loss": 0.00010728836059570312,
"retain_loss": 5.1021575927734375e-05,
"step": 130
},
{
"epoch": 0.48826291079812206,
"forget_loss": 0.00010204315185546875,
"retain_loss": 0.00011110305786132812,
"step": 130
},
{
"epoch": 0.48826291079812206,
"forget_loss": 0.000156402587890625,
"retain_loss": 7.05718994140625e-05,
"step": 130
},
{
"epoch": 0.5258215962441315,
"grad_norm": 0.035400390625,
"learning_rate": 7.684824902723736e-06,
"loss": 0.000272369384765625,
"step": 140
},
{
"epoch": 0.5258215962441315,
"forget_loss": 0.0001087188720703125,
"retain_loss": 7.009506225585938e-05,
"step": 140
},
{
"epoch": 0.5258215962441315,
"forget_loss": 0.0001239776611328125,
"retain_loss": 7.963180541992188e-05,
"step": 140
},
{
"epoch": 0.5258215962441315,
"forget_loss": 0.00017070770263671875,
"retain_loss": 6.29425048828125e-05,
"step": 140
},
{
"epoch": 0.5258215962441315,
"forget_loss": 0.00013446807861328125,
"retain_loss": 5.5789947509765625e-05,
"step": 140
},
{
"epoch": 0.5633802816901409,
"grad_norm": 0.0177001953125,
"learning_rate": 7.4902723735408565e-06,
"loss": 0.00019230842590332032,
"step": 150
},
{
"epoch": 0.5633802816901409,
"forget_loss": 0.00010585784912109375,
"retain_loss": 5.793571472167969e-05,
"step": 150
},
{
"epoch": 0.5633802816901409,
"forget_loss": 0.00011157989501953125,
"retain_loss": 0.00015163421630859375,
"step": 150
},
{
"epoch": 0.5633802816901409,
"forget_loss": 8.821487426757812e-05,
"retain_loss": 6.866455078125e-05,
"step": 150
},
{
"epoch": 0.5633802816901409,
"forget_loss": 9.059906005859375e-05,
"retain_loss": 0.00010251998901367188,
"step": 150
},
{
"epoch": 0.6009389671361502,
"grad_norm": 0.06591796875,
"learning_rate": 7.295719844357977e-06,
"loss": 0.00020923614501953126,
"step": 160
},
{
"epoch": 0.6009389671361502,
"forget_loss": 0.00010156631469726562,
"retain_loss": 8.535385131835938e-05,
"step": 160
},
{
"epoch": 0.6009389671361502,
"forget_loss": 9.441375732421875e-05,
"retain_loss": 7.05718994140625e-05,
"step": 160
},
{
"epoch": 0.6009389671361502,
"forget_loss": 9.489059448242188e-05,
"retain_loss": 6.151199340820312e-05,
"step": 160
},
{
"epoch": 0.6009389671361502,
"forget_loss": 9.679794311523438e-05,
"retain_loss": 5.984306335449219e-05,
"step": 160
},
{
"epoch": 0.6384976525821596,
"grad_norm": 0.0152587890625,
"learning_rate": 7.101167315175098e-06,
"loss": 0.0001931428909301758,
"step": 170
},
{
"epoch": 0.6384976525821596,
"forget_loss": 9.489059448242188e-05,
"retain_loss": 6.079673767089844e-05,
"step": 170
},
{
"epoch": 0.6384976525821596,
"forget_loss": 0.000110626220703125,
"retain_loss": 7.104873657226562e-05,
"step": 170
},
{
"epoch": 0.6384976525821596,
"forget_loss": 0.00010776519775390625,
"retain_loss": 0.00011539459228515625,
"step": 170
},
{
"epoch": 0.6384976525821596,
"forget_loss": 9.441375732421875e-05,
"retain_loss": 0.00011682510375976562,
"step": 170
},
{
"epoch": 0.676056338028169,
"grad_norm": 0.0306396484375,
"learning_rate": 6.9066147859922184e-06,
"loss": 0.00021104812622070314,
"step": 180
},
{
"epoch": 0.676056338028169,
"forget_loss": 0.000125885009765625,
"retain_loss": 0.0001239776611328125,
"step": 180
},
{
"epoch": 0.676056338028169,
"forget_loss": 0.00010395050048828125,
"retain_loss": 8.96453857421875e-05,
"step": 180
},
{
"epoch": 0.676056338028169,
"forget_loss": 8.869171142578125e-05,
"retain_loss": 8.20159912109375e-05,
"step": 180
},
{
"epoch": 0.676056338028169,
"forget_loss": 9.107589721679688e-05,
"retain_loss": 5.316734313964844e-05,
"step": 180
},
{
"epoch": 0.7136150234741784,
"grad_norm": 0.0230712890625,
"learning_rate": 6.712062256809338e-06,
"loss": 0.0001739978790283203,
"step": 190
},
{
"epoch": 0.7136150234741784,
"forget_loss": 9.059906005859375e-05,
"retain_loss": 5.793571472167969e-05,
"step": 190
},
{
"epoch": 0.7136150234741784,
"forget_loss": 9.5367431640625e-05,
"retain_loss": 5.0067901611328125e-05,
"step": 190
},
{
"epoch": 0.7136150234741784,
"forget_loss": 8.869171142578125e-05,
"retain_loss": 5.7220458984375e-05,
"step": 190
},
{
"epoch": 0.7136150234741784,
"forget_loss": 0.00010919570922851562,
"retain_loss": 4.9591064453125e-05,
"step": 190
},
{
"epoch": 0.7511737089201878,
"grad_norm": 0.01068115234375,
"learning_rate": 6.51750972762646e-06,
"loss": 0.00019364356994628907,
"step": 200
}
],
"logging_steps": 10,
"max_steps": 534,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 200,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 0.0,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}