Training in progress, step 12000, checkpoint
Browse files- last-checkpoint/optimizer.pt +1 -1
- last-checkpoint/pytorch_model.bin +1 -1
- last-checkpoint/rng_state_0.pth +1 -1
- last-checkpoint/rng_state_1.pth +1 -1
- last-checkpoint/rng_state_2.pth +1 -1
- last-checkpoint/rng_state_3.pth +1 -1
- last-checkpoint/scheduler.pt +1 -1
- last-checkpoint/trainer_state.json +353 -3
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 244223098
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:53a76ac9077fe4281340d991c88d870e9612c562f18291ef2fff16717ae5fa8e
|
| 3 |
size 244223098
|
last-checkpoint/pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 381944306
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:91d1f43ad07d5d2506105a4a30a104f55de4d303f90118c0693725602ca12996
|
| 3 |
size 381944306
|
last-checkpoint/rng_state_0.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14960
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9ad09e3903383f043ca34c37e900cc005c86f2fe664b64f1c40d71b64081bc5f
|
| 3 |
size 14960
|
last-checkpoint/rng_state_1.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14960
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:394ea9776e43b9e34e0f31ed80a4339ee1363ca5d97e2a122bc469ea88e3a051
|
| 3 |
size 14960
|
last-checkpoint/rng_state_2.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14960
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:185d9bf5dd99d2fb4439e3538848f269528414d649a4a521e708b152f704c300
|
| 3 |
size 14960
|
last-checkpoint/rng_state_3.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14960
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f8af41d2465acba77e73ddefdb8cd3bcfad2e25c1dc9b4e154733802fd9eb85a
|
| 3 |
size 14960
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c91c39eca7bff6b168c2d221fc52c0c253070df5bf1f21ee503aeddc471ee587
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -2,9 +2,9 @@
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 0.
|
| 6 |
"eval_steps": 500,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -3858,6 +3858,356 @@
|
|
| 3858 |
"learning_rate": 0.0004974073060677781,
|
| 3859 |
"loss": 23.6657,
|
| 3860 |
"step": 11000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3861 |
}
|
| 3862 |
],
|
| 3863 |
"logging_steps": 20,
|
|
@@ -3877,7 +4227,7 @@
|
|
| 3877 |
"attributes": {}
|
| 3878 |
}
|
| 3879 |
},
|
| 3880 |
-
"total_flos":
|
| 3881 |
"train_batch_size": 48,
|
| 3882 |
"trial_name": null,
|
| 3883 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.017775776357032393,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 12000,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 3858 |
"learning_rate": 0.0004974073060677781,
|
| 3859 |
"loss": 23.6657,
|
| 3860 |
"step": 11000
|
| 3861 |
+
},
|
| 3862 |
+
{
|
| 3863 |
+
"epoch": 0.016324087954541416,
|
| 3864 |
+
"grad_norm": 18.875,
|
| 3865 |
+
"learning_rate": 0.0004974023671327705,
|
| 3866 |
+
"loss": 23.6059,
|
| 3867 |
+
"step": 11020
|
| 3868 |
+
},
|
| 3869 |
+
{
|
| 3870 |
+
"epoch": 0.016353714248469802,
|
| 3871 |
+
"grad_norm": 16.875,
|
| 3872 |
+
"learning_rate": 0.0004973974281977628,
|
| 3873 |
+
"loss": 23.6203,
|
| 3874 |
+
"step": 11040
|
| 3875 |
+
},
|
| 3876 |
+
{
|
| 3877 |
+
"epoch": 0.01638334054239819,
|
| 3878 |
+
"grad_norm": 26.0,
|
| 3879 |
+
"learning_rate": 0.0004973924892627553,
|
| 3880 |
+
"loss": 23.5207,
|
| 3881 |
+
"step": 11060
|
| 3882 |
+
},
|
| 3883 |
+
{
|
| 3884 |
+
"epoch": 0.016412966836326575,
|
| 3885 |
+
"grad_norm": 18.25,
|
| 3886 |
+
"learning_rate": 0.0004973875503277477,
|
| 3887 |
+
"loss": 23.711,
|
| 3888 |
+
"step": 11080
|
| 3889 |
+
},
|
| 3890 |
+
{
|
| 3891 |
+
"epoch": 0.016442593130254965,
|
| 3892 |
+
"grad_norm": 17.125,
|
| 3893 |
+
"learning_rate": 0.0004973826113927401,
|
| 3894 |
+
"loss": 23.5764,
|
| 3895 |
+
"step": 11100
|
| 3896 |
+
},
|
| 3897 |
+
{
|
| 3898 |
+
"epoch": 0.01647221942418335,
|
| 3899 |
+
"grad_norm": 18.125,
|
| 3900 |
+
"learning_rate": 0.0004973776724577325,
|
| 3901 |
+
"loss": 23.6693,
|
| 3902 |
+
"step": 11120
|
| 3903 |
+
},
|
| 3904 |
+
{
|
| 3905 |
+
"epoch": 0.016501845718111738,
|
| 3906 |
+
"grad_norm": 20.875,
|
| 3907 |
+
"learning_rate": 0.000497372733522725,
|
| 3908 |
+
"loss": 23.5375,
|
| 3909 |
+
"step": 11140
|
| 3910 |
+
},
|
| 3911 |
+
{
|
| 3912 |
+
"epoch": 0.016531472012040124,
|
| 3913 |
+
"grad_norm": 14.75,
|
| 3914 |
+
"learning_rate": 0.0004973677945877174,
|
| 3915 |
+
"loss": 23.5473,
|
| 3916 |
+
"step": 11160
|
| 3917 |
+
},
|
| 3918 |
+
{
|
| 3919 |
+
"epoch": 0.016561098305968514,
|
| 3920 |
+
"grad_norm": 15.625,
|
| 3921 |
+
"learning_rate": 0.0004973628556527099,
|
| 3922 |
+
"loss": 23.5889,
|
| 3923 |
+
"step": 11180
|
| 3924 |
+
},
|
| 3925 |
+
{
|
| 3926 |
+
"epoch": 0.0165907245998969,
|
| 3927 |
+
"grad_norm": 16.875,
|
| 3928 |
+
"learning_rate": 0.0004973579167177023,
|
| 3929 |
+
"loss": 23.5879,
|
| 3930 |
+
"step": 11200
|
| 3931 |
+
},
|
| 3932 |
+
{
|
| 3933 |
+
"epoch": 0.016620350893825287,
|
| 3934 |
+
"grad_norm": 18.375,
|
| 3935 |
+
"learning_rate": 0.0004973529777826948,
|
| 3936 |
+
"loss": 23.4974,
|
| 3937 |
+
"step": 11220
|
| 3938 |
+
},
|
| 3939 |
+
{
|
| 3940 |
+
"epoch": 0.016649977187753674,
|
| 3941 |
+
"grad_norm": 15.625,
|
| 3942 |
+
"learning_rate": 0.0004973480388476872,
|
| 3943 |
+
"loss": 23.4771,
|
| 3944 |
+
"step": 11240
|
| 3945 |
+
},
|
| 3946 |
+
{
|
| 3947 |
+
"epoch": 0.016679603481682063,
|
| 3948 |
+
"grad_norm": 17.5,
|
| 3949 |
+
"learning_rate": 0.0004973430999126796,
|
| 3950 |
+
"loss": 23.4806,
|
| 3951 |
+
"step": 11260
|
| 3952 |
+
},
|
| 3953 |
+
{
|
| 3954 |
+
"epoch": 0.01670922977561045,
|
| 3955 |
+
"grad_norm": 19.75,
|
| 3956 |
+
"learning_rate": 0.0004973381609776721,
|
| 3957 |
+
"loss": 23.651,
|
| 3958 |
+
"step": 11280
|
| 3959 |
+
},
|
| 3960 |
+
{
|
| 3961 |
+
"epoch": 0.016738856069538836,
|
| 3962 |
+
"grad_norm": 16.625,
|
| 3963 |
+
"learning_rate": 0.0004973332220426645,
|
| 3964 |
+
"loss": 23.5367,
|
| 3965 |
+
"step": 11300
|
| 3966 |
+
},
|
| 3967 |
+
{
|
| 3968 |
+
"epoch": 0.016768482363467223,
|
| 3969 |
+
"grad_norm": 19.875,
|
| 3970 |
+
"learning_rate": 0.0004973282831076569,
|
| 3971 |
+
"loss": 23.5171,
|
| 3972 |
+
"step": 11320
|
| 3973 |
+
},
|
| 3974 |
+
{
|
| 3975 |
+
"epoch": 0.016798108657395613,
|
| 3976 |
+
"grad_norm": 17.125,
|
| 3977 |
+
"learning_rate": 0.0004973233441726494,
|
| 3978 |
+
"loss": 23.4766,
|
| 3979 |
+
"step": 11340
|
| 3980 |
+
},
|
| 3981 |
+
{
|
| 3982 |
+
"epoch": 0.016827734951324,
|
| 3983 |
+
"grad_norm": 15.3125,
|
| 3984 |
+
"learning_rate": 0.0004973184052376418,
|
| 3985 |
+
"loss": 23.4622,
|
| 3986 |
+
"step": 11360
|
| 3987 |
+
},
|
| 3988 |
+
{
|
| 3989 |
+
"epoch": 0.016857361245252386,
|
| 3990 |
+
"grad_norm": 19.375,
|
| 3991 |
+
"learning_rate": 0.0004973134663026343,
|
| 3992 |
+
"loss": 23.5135,
|
| 3993 |
+
"step": 11380
|
| 3994 |
+
},
|
| 3995 |
+
{
|
| 3996 |
+
"epoch": 0.016886987539180772,
|
| 3997 |
+
"grad_norm": 17.625,
|
| 3998 |
+
"learning_rate": 0.0004973085273676267,
|
| 3999 |
+
"loss": 23.485,
|
| 4000 |
+
"step": 11400
|
| 4001 |
+
},
|
| 4002 |
+
{
|
| 4003 |
+
"epoch": 0.016916613833109162,
|
| 4004 |
+
"grad_norm": 17.5,
|
| 4005 |
+
"learning_rate": 0.0004973035884326191,
|
| 4006 |
+
"loss": 23.4218,
|
| 4007 |
+
"step": 11420
|
| 4008 |
+
},
|
| 4009 |
+
{
|
| 4010 |
+
"epoch": 0.01694624012703755,
|
| 4011 |
+
"grad_norm": 16.375,
|
| 4012 |
+
"learning_rate": 0.0004972986494976116,
|
| 4013 |
+
"loss": 23.4405,
|
| 4014 |
+
"step": 11440
|
| 4015 |
+
},
|
| 4016 |
+
{
|
| 4017 |
+
"epoch": 0.016975866420965935,
|
| 4018 |
+
"grad_norm": 21.125,
|
| 4019 |
+
"learning_rate": 0.000497293710562604,
|
| 4020 |
+
"loss": 23.4308,
|
| 4021 |
+
"step": 11460
|
| 4022 |
+
},
|
| 4023 |
+
{
|
| 4024 |
+
"epoch": 0.01700549271489432,
|
| 4025 |
+
"grad_norm": 18.875,
|
| 4026 |
+
"learning_rate": 0.0004972887716275964,
|
| 4027 |
+
"loss": 23.4076,
|
| 4028 |
+
"step": 11480
|
| 4029 |
+
},
|
| 4030 |
+
{
|
| 4031 |
+
"epoch": 0.01703511900882271,
|
| 4032 |
+
"grad_norm": 16.25,
|
| 4033 |
+
"learning_rate": 0.0004972838326925889,
|
| 4034 |
+
"loss": 23.4027,
|
| 4035 |
+
"step": 11500
|
| 4036 |
+
},
|
| 4037 |
+
{
|
| 4038 |
+
"epoch": 0.017064745302751098,
|
| 4039 |
+
"grad_norm": 15.9375,
|
| 4040 |
+
"learning_rate": 0.0004972788937575813,
|
| 4041 |
+
"loss": 23.3797,
|
| 4042 |
+
"step": 11520
|
| 4043 |
+
},
|
| 4044 |
+
{
|
| 4045 |
+
"epoch": 0.017094371596679484,
|
| 4046 |
+
"grad_norm": 17.875,
|
| 4047 |
+
"learning_rate": 0.0004972739548225737,
|
| 4048 |
+
"loss": 23.403,
|
| 4049 |
+
"step": 11540
|
| 4050 |
+
},
|
| 4051 |
+
{
|
| 4052 |
+
"epoch": 0.01712399789060787,
|
| 4053 |
+
"grad_norm": 20.25,
|
| 4054 |
+
"learning_rate": 0.0004972690158875662,
|
| 4055 |
+
"loss": 23.435,
|
| 4056 |
+
"step": 11560
|
| 4057 |
+
},
|
| 4058 |
+
{
|
| 4059 |
+
"epoch": 0.01715362418453626,
|
| 4060 |
+
"grad_norm": 17.875,
|
| 4061 |
+
"learning_rate": 0.0004972640769525586,
|
| 4062 |
+
"loss": 23.3429,
|
| 4063 |
+
"step": 11580
|
| 4064 |
+
},
|
| 4065 |
+
{
|
| 4066 |
+
"epoch": 0.017183250478464647,
|
| 4067 |
+
"grad_norm": 19.0,
|
| 4068 |
+
"learning_rate": 0.0004972591380175511,
|
| 4069 |
+
"loss": 23.3458,
|
| 4070 |
+
"step": 11600
|
| 4071 |
+
},
|
| 4072 |
+
{
|
| 4073 |
+
"epoch": 0.017212876772393033,
|
| 4074 |
+
"grad_norm": 14.3125,
|
| 4075 |
+
"learning_rate": 0.0004972541990825435,
|
| 4076 |
+
"loss": 23.3765,
|
| 4077 |
+
"step": 11620
|
| 4078 |
+
},
|
| 4079 |
+
{
|
| 4080 |
+
"epoch": 0.01724250306632142,
|
| 4081 |
+
"grad_norm": 17.375,
|
| 4082 |
+
"learning_rate": 0.0004972492601475358,
|
| 4083 |
+
"loss": 23.3812,
|
| 4084 |
+
"step": 11640
|
| 4085 |
+
},
|
| 4086 |
+
{
|
| 4087 |
+
"epoch": 0.01727212936024981,
|
| 4088 |
+
"grad_norm": 17.375,
|
| 4089 |
+
"learning_rate": 0.0004972443212125283,
|
| 4090 |
+
"loss": 23.3419,
|
| 4091 |
+
"step": 11660
|
| 4092 |
+
},
|
| 4093 |
+
{
|
| 4094 |
+
"epoch": 0.017301755654178196,
|
| 4095 |
+
"grad_norm": 15.9375,
|
| 4096 |
+
"learning_rate": 0.0004972393822775207,
|
| 4097 |
+
"loss": 23.1804,
|
| 4098 |
+
"step": 11680
|
| 4099 |
+
},
|
| 4100 |
+
{
|
| 4101 |
+
"epoch": 0.017331381948106583,
|
| 4102 |
+
"grad_norm": 15.125,
|
| 4103 |
+
"learning_rate": 0.0004972344433425131,
|
| 4104 |
+
"loss": 23.2947,
|
| 4105 |
+
"step": 11700
|
| 4106 |
+
},
|
| 4107 |
+
{
|
| 4108 |
+
"epoch": 0.01736100824203497,
|
| 4109 |
+
"grad_norm": 17.0,
|
| 4110 |
+
"learning_rate": 0.0004972295044075056,
|
| 4111 |
+
"loss": 23.402,
|
| 4112 |
+
"step": 11720
|
| 4113 |
+
},
|
| 4114 |
+
{
|
| 4115 |
+
"epoch": 0.01739063453596336,
|
| 4116 |
+
"grad_norm": 18.5,
|
| 4117 |
+
"learning_rate": 0.000497224565472498,
|
| 4118 |
+
"loss": 23.2933,
|
| 4119 |
+
"step": 11740
|
| 4120 |
+
},
|
| 4121 |
+
{
|
| 4122 |
+
"epoch": 0.017420260829891746,
|
| 4123 |
+
"grad_norm": 17.125,
|
| 4124 |
+
"learning_rate": 0.0004972196265374904,
|
| 4125 |
+
"loss": 23.296,
|
| 4126 |
+
"step": 11760
|
| 4127 |
+
},
|
| 4128 |
+
{
|
| 4129 |
+
"epoch": 0.017449887123820132,
|
| 4130 |
+
"grad_norm": 15.5625,
|
| 4131 |
+
"learning_rate": 0.0004972146876024829,
|
| 4132 |
+
"loss": 23.2247,
|
| 4133 |
+
"step": 11780
|
| 4134 |
+
},
|
| 4135 |
+
{
|
| 4136 |
+
"epoch": 0.017479513417748522,
|
| 4137 |
+
"grad_norm": 15.375,
|
| 4138 |
+
"learning_rate": 0.0004972097486674753,
|
| 4139 |
+
"loss": 23.1945,
|
| 4140 |
+
"step": 11800
|
| 4141 |
+
},
|
| 4142 |
+
{
|
| 4143 |
+
"epoch": 0.01750913971167691,
|
| 4144 |
+
"grad_norm": 17.25,
|
| 4145 |
+
"learning_rate": 0.0004972048097324678,
|
| 4146 |
+
"loss": 23.2879,
|
| 4147 |
+
"step": 11820
|
| 4148 |
+
},
|
| 4149 |
+
{
|
| 4150 |
+
"epoch": 0.017538766005605295,
|
| 4151 |
+
"grad_norm": 16.5,
|
| 4152 |
+
"learning_rate": 0.0004971998707974602,
|
| 4153 |
+
"loss": 23.2503,
|
| 4154 |
+
"step": 11840
|
| 4155 |
+
},
|
| 4156 |
+
{
|
| 4157 |
+
"epoch": 0.01756839229953368,
|
| 4158 |
+
"grad_norm": 16.875,
|
| 4159 |
+
"learning_rate": 0.0004971949318624526,
|
| 4160 |
+
"loss": 23.2298,
|
| 4161 |
+
"step": 11860
|
| 4162 |
+
},
|
| 4163 |
+
{
|
| 4164 |
+
"epoch": 0.01759801859346207,
|
| 4165 |
+
"grad_norm": 16.5,
|
| 4166 |
+
"learning_rate": 0.0004971899929274451,
|
| 4167 |
+
"loss": 23.2478,
|
| 4168 |
+
"step": 11880
|
| 4169 |
+
},
|
| 4170 |
+
{
|
| 4171 |
+
"epoch": 0.017627644887390458,
|
| 4172 |
+
"grad_norm": 16.875,
|
| 4173 |
+
"learning_rate": 0.0004971850539924375,
|
| 4174 |
+
"loss": 23.2439,
|
| 4175 |
+
"step": 11900
|
| 4176 |
+
},
|
| 4177 |
+
{
|
| 4178 |
+
"epoch": 0.017657271181318844,
|
| 4179 |
+
"grad_norm": 16.75,
|
| 4180 |
+
"learning_rate": 0.0004971801150574299,
|
| 4181 |
+
"loss": 23.2426,
|
| 4182 |
+
"step": 11920
|
| 4183 |
+
},
|
| 4184 |
+
{
|
| 4185 |
+
"epoch": 0.01768689747524723,
|
| 4186 |
+
"grad_norm": 15.5,
|
| 4187 |
+
"learning_rate": 0.0004971751761224224,
|
| 4188 |
+
"loss": 23.2587,
|
| 4189 |
+
"step": 11940
|
| 4190 |
+
},
|
| 4191 |
+
{
|
| 4192 |
+
"epoch": 0.01771652376917562,
|
| 4193 |
+
"grad_norm": 15.6875,
|
| 4194 |
+
"learning_rate": 0.0004971702371874148,
|
| 4195 |
+
"loss": 23.2458,
|
| 4196 |
+
"step": 11960
|
| 4197 |
+
},
|
| 4198 |
+
{
|
| 4199 |
+
"epoch": 0.017746150063104007,
|
| 4200 |
+
"grad_norm": 17.875,
|
| 4201 |
+
"learning_rate": 0.0004971652982524073,
|
| 4202 |
+
"loss": 23.1944,
|
| 4203 |
+
"step": 11980
|
| 4204 |
+
},
|
| 4205 |
+
{
|
| 4206 |
+
"epoch": 0.017775776357032393,
|
| 4207 |
+
"grad_norm": 15.5,
|
| 4208 |
+
"learning_rate": 0.0004971603593173997,
|
| 4209 |
+
"loss": 23.2169,
|
| 4210 |
+
"step": 12000
|
| 4211 |
}
|
| 4212 |
],
|
| 4213 |
"logging_steps": 20,
|
|
|
|
| 4227 |
"attributes": {}
|
| 4228 |
}
|
| 4229 |
},
|
| 4230 |
+
"total_flos": 8.255909218322743e+18,
|
| 4231 |
"train_batch_size": 48,
|
| 4232 |
"trial_name": null,
|
| 4233 |
"trial_params": null
|