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 319352826
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fe4d66c1c45da3d7c1f12fac8784e3a9ed3f9c80aaaeb6d876b0d9ad695ad6ac
|
| 3 |
size 319352826
|
last-checkpoint/pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 900372486
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:037cbabeb904fe4bc6ddc6347802807d0fa4aee83b97d95773c6ab9e0710b4c5
|
| 3 |
size 900372486
|
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 sha256:0f66d2cdb2727898aa57d1805f0959aa2d5b4d00101c612d89b884ee5f99403d
|
| 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 sha256:1c45c8627a787157525e0441f74e851dbeed0484ecee31ac2a6d27081cbd4784
|
| 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 sha256:668b78f9e33a1b391d4ae9971e15c64123216cc17e0701f0c88ed161b79d182b
|
| 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 sha256:3a22c853237523fba49f3eb78490cfffdd83e225ddbf7f5581f622db7e7b1ed5
|
| 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 sha256:75ee9351d0b9f6c92b7d23f6bec3e5476b4b37acddcbc9d54675b0b4f4ee65b1
|
| 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": 4.97407306067778e-05,
|
| 3859 |
"loss": 25.0612,
|
| 3860 |
"step": 11000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3861 |
}
|
| 3862 |
],
|
| 3863 |
"logging_steps": 20,
|
|
@@ -3877,7 +4227,7 @@
|
|
| 3877 |
"attributes": {}
|
| 3878 |
}
|
| 3879 |
},
|
| 3880 |
-
"total_flos": 2.
|
| 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": 4.97407306067778e-05,
|
| 3859 |
"loss": 25.0612,
|
| 3860 |
"step": 11000
|
| 3861 |
+
},
|
| 3862 |
+
{
|
| 3863 |
+
"epoch": 0.016324087954541416,
|
| 3864 |
+
"grad_norm": 8.875,
|
| 3865 |
+
"learning_rate": 4.9740236713277047e-05,
|
| 3866 |
+
"loss": 24.9356,
|
| 3867 |
+
"step": 11020
|
| 3868 |
+
},
|
| 3869 |
+
{
|
| 3870 |
+
"epoch": 0.016353714248469802,
|
| 3871 |
+
"grad_norm": 8.9375,
|
| 3872 |
+
"learning_rate": 4.973974281977628e-05,
|
| 3873 |
+
"loss": 24.9984,
|
| 3874 |
+
"step": 11040
|
| 3875 |
+
},
|
| 3876 |
+
{
|
| 3877 |
+
"epoch": 0.01638334054239819,
|
| 3878 |
+
"grad_norm": 7.46875,
|
| 3879 |
+
"learning_rate": 4.9739248926275533e-05,
|
| 3880 |
+
"loss": 24.9063,
|
| 3881 |
+
"step": 11060
|
| 3882 |
+
},
|
| 3883 |
+
{
|
| 3884 |
+
"epoch": 0.016412966836326575,
|
| 3885 |
+
"grad_norm": 7.125,
|
| 3886 |
+
"learning_rate": 4.973875503277478e-05,
|
| 3887 |
+
"loss": 25.0004,
|
| 3888 |
+
"step": 11080
|
| 3889 |
+
},
|
| 3890 |
+
{
|
| 3891 |
+
"epoch": 0.016442593130254965,
|
| 3892 |
+
"grad_norm": 8.6875,
|
| 3893 |
+
"learning_rate": 4.973826113927402e-05,
|
| 3894 |
+
"loss": 24.8795,
|
| 3895 |
+
"step": 11100
|
| 3896 |
+
},
|
| 3897 |
+
{
|
| 3898 |
+
"epoch": 0.01647221942418335,
|
| 3899 |
+
"grad_norm": 7.28125,
|
| 3900 |
+
"learning_rate": 4.973776724577326e-05,
|
| 3901 |
+
"loss": 24.9027,
|
| 3902 |
+
"step": 11120
|
| 3903 |
+
},
|
| 3904 |
+
{
|
| 3905 |
+
"epoch": 0.016501845718111738,
|
| 3906 |
+
"grad_norm": 7.53125,
|
| 3907 |
+
"learning_rate": 4.973727335227251e-05,
|
| 3908 |
+
"loss": 24.885,
|
| 3909 |
+
"step": 11140
|
| 3910 |
+
},
|
| 3911 |
+
{
|
| 3912 |
+
"epoch": 0.016531472012040124,
|
| 3913 |
+
"grad_norm": 7.8125,
|
| 3914 |
+
"learning_rate": 4.973677945877175e-05,
|
| 3915 |
+
"loss": 24.9615,
|
| 3916 |
+
"step": 11160
|
| 3917 |
+
},
|
| 3918 |
+
{
|
| 3919 |
+
"epoch": 0.016561098305968514,
|
| 3920 |
+
"grad_norm": 9.125,
|
| 3921 |
+
"learning_rate": 4.973628556527099e-05,
|
| 3922 |
+
"loss": 25.0393,
|
| 3923 |
+
"step": 11180
|
| 3924 |
+
},
|
| 3925 |
+
{
|
| 3926 |
+
"epoch": 0.0165907245998969,
|
| 3927 |
+
"grad_norm": 9.5625,
|
| 3928 |
+
"learning_rate": 4.973579167177024e-05,
|
| 3929 |
+
"loss": 25.0047,
|
| 3930 |
+
"step": 11200
|
| 3931 |
+
},
|
| 3932 |
+
{
|
| 3933 |
+
"epoch": 0.016620350893825287,
|
| 3934 |
+
"grad_norm": 7.6875,
|
| 3935 |
+
"learning_rate": 4.973529777826948e-05,
|
| 3936 |
+
"loss": 24.8962,
|
| 3937 |
+
"step": 11220
|
| 3938 |
+
},
|
| 3939 |
+
{
|
| 3940 |
+
"epoch": 0.016649977187753674,
|
| 3941 |
+
"grad_norm": 8.5,
|
| 3942 |
+
"learning_rate": 4.9734803884768724e-05,
|
| 3943 |
+
"loss": 24.9105,
|
| 3944 |
+
"step": 11240
|
| 3945 |
+
},
|
| 3946 |
+
{
|
| 3947 |
+
"epoch": 0.016679603481682063,
|
| 3948 |
+
"grad_norm": 8.375,
|
| 3949 |
+
"learning_rate": 4.973430999126796e-05,
|
| 3950 |
+
"loss": 24.9264,
|
| 3951 |
+
"step": 11260
|
| 3952 |
+
},
|
| 3953 |
+
{
|
| 3954 |
+
"epoch": 0.01670922977561045,
|
| 3955 |
+
"grad_norm": 11.625,
|
| 3956 |
+
"learning_rate": 4.973381609776721e-05,
|
| 3957 |
+
"loss": 24.9545,
|
| 3958 |
+
"step": 11280
|
| 3959 |
+
},
|
| 3960 |
+
{
|
| 3961 |
+
"epoch": 0.016738856069538836,
|
| 3962 |
+
"grad_norm": 6.875,
|
| 3963 |
+
"learning_rate": 4.9733322204266455e-05,
|
| 3964 |
+
"loss": 24.8835,
|
| 3965 |
+
"step": 11300
|
| 3966 |
+
},
|
| 3967 |
+
{
|
| 3968 |
+
"epoch": 0.016768482363467223,
|
| 3969 |
+
"grad_norm": 7.625,
|
| 3970 |
+
"learning_rate": 4.973282831076569e-05,
|
| 3971 |
+
"loss": 24.8933,
|
| 3972 |
+
"step": 11320
|
| 3973 |
+
},
|
| 3974 |
+
{
|
| 3975 |
+
"epoch": 0.016798108657395613,
|
| 3976 |
+
"grad_norm": 8.375,
|
| 3977 |
+
"learning_rate": 4.9732334417264935e-05,
|
| 3978 |
+
"loss": 24.9687,
|
| 3979 |
+
"step": 11340
|
| 3980 |
+
},
|
| 3981 |
+
{
|
| 3982 |
+
"epoch": 0.016827734951324,
|
| 3983 |
+
"grad_norm": 7.65625,
|
| 3984 |
+
"learning_rate": 4.9731840523764185e-05,
|
| 3985 |
+
"loss": 24.9447,
|
| 3986 |
+
"step": 11360
|
| 3987 |
+
},
|
| 3988 |
+
{
|
| 3989 |
+
"epoch": 0.016857361245252386,
|
| 3990 |
+
"grad_norm": 7.46875,
|
| 3991 |
+
"learning_rate": 4.973134663026343e-05,
|
| 3992 |
+
"loss": 24.8808,
|
| 3993 |
+
"step": 11380
|
| 3994 |
+
},
|
| 3995 |
+
{
|
| 3996 |
+
"epoch": 0.016886987539180772,
|
| 3997 |
+
"grad_norm": 9.125,
|
| 3998 |
+
"learning_rate": 4.9730852736762665e-05,
|
| 3999 |
+
"loss": 24.931,
|
| 4000 |
+
"step": 11400
|
| 4001 |
+
},
|
| 4002 |
+
{
|
| 4003 |
+
"epoch": 0.016916613833109162,
|
| 4004 |
+
"grad_norm": 9.5625,
|
| 4005 |
+
"learning_rate": 4.973035884326191e-05,
|
| 4006 |
+
"loss": 24.9442,
|
| 4007 |
+
"step": 11420
|
| 4008 |
+
},
|
| 4009 |
+
{
|
| 4010 |
+
"epoch": 0.01694624012703755,
|
| 4011 |
+
"grad_norm": 7.09375,
|
| 4012 |
+
"learning_rate": 4.972986494976116e-05,
|
| 4013 |
+
"loss": 24.8634,
|
| 4014 |
+
"step": 11440
|
| 4015 |
+
},
|
| 4016 |
+
{
|
| 4017 |
+
"epoch": 0.016975866420965935,
|
| 4018 |
+
"grad_norm": 7.71875,
|
| 4019 |
+
"learning_rate": 4.97293710562604e-05,
|
| 4020 |
+
"loss": 24.9538,
|
| 4021 |
+
"step": 11460
|
| 4022 |
+
},
|
| 4023 |
+
{
|
| 4024 |
+
"epoch": 0.01700549271489432,
|
| 4025 |
+
"grad_norm": 8.3125,
|
| 4026 |
+
"learning_rate": 4.972887716275964e-05,
|
| 4027 |
+
"loss": 24.8944,
|
| 4028 |
+
"step": 11480
|
| 4029 |
+
},
|
| 4030 |
+
{
|
| 4031 |
+
"epoch": 0.01703511900882271,
|
| 4032 |
+
"grad_norm": 11.25,
|
| 4033 |
+
"learning_rate": 4.972838326925889e-05,
|
| 4034 |
+
"loss": 24.8674,
|
| 4035 |
+
"step": 11500
|
| 4036 |
+
},
|
| 4037 |
+
{
|
| 4038 |
+
"epoch": 0.017064745302751098,
|
| 4039 |
+
"grad_norm": 7.21875,
|
| 4040 |
+
"learning_rate": 4.972788937575813e-05,
|
| 4041 |
+
"loss": 24.9174,
|
| 4042 |
+
"step": 11520
|
| 4043 |
+
},
|
| 4044 |
+
{
|
| 4045 |
+
"epoch": 0.017094371596679484,
|
| 4046 |
+
"grad_norm": 8.1875,
|
| 4047 |
+
"learning_rate": 4.972739548225737e-05,
|
| 4048 |
+
"loss": 24.8759,
|
| 4049 |
+
"step": 11540
|
| 4050 |
+
},
|
| 4051 |
+
{
|
| 4052 |
+
"epoch": 0.01712399789060787,
|
| 4053 |
+
"grad_norm": 10.0625,
|
| 4054 |
+
"learning_rate": 4.972690158875661e-05,
|
| 4055 |
+
"loss": 24.8022,
|
| 4056 |
+
"step": 11560
|
| 4057 |
+
},
|
| 4058 |
+
{
|
| 4059 |
+
"epoch": 0.01715362418453626,
|
| 4060 |
+
"grad_norm": 9.4375,
|
| 4061 |
+
"learning_rate": 4.972640769525586e-05,
|
| 4062 |
+
"loss": 24.8826,
|
| 4063 |
+
"step": 11580
|
| 4064 |
+
},
|
| 4065 |
+
{
|
| 4066 |
+
"epoch": 0.017183250478464647,
|
| 4067 |
+
"grad_norm": 9.0625,
|
| 4068 |
+
"learning_rate": 4.9725913801755106e-05,
|
| 4069 |
+
"loss": 24.8848,
|
| 4070 |
+
"step": 11600
|
| 4071 |
+
},
|
| 4072 |
+
{
|
| 4073 |
+
"epoch": 0.017212876772393033,
|
| 4074 |
+
"grad_norm": 9.4375,
|
| 4075 |
+
"learning_rate": 4.972541990825434e-05,
|
| 4076 |
+
"loss": 24.7912,
|
| 4077 |
+
"step": 11620
|
| 4078 |
+
},
|
| 4079 |
+
{
|
| 4080 |
+
"epoch": 0.01724250306632142,
|
| 4081 |
+
"grad_norm": 8.875,
|
| 4082 |
+
"learning_rate": 4.9724926014753586e-05,
|
| 4083 |
+
"loss": 24.9131,
|
| 4084 |
+
"step": 11640
|
| 4085 |
+
},
|
| 4086 |
+
{
|
| 4087 |
+
"epoch": 0.01727212936024981,
|
| 4088 |
+
"grad_norm": 9.3125,
|
| 4089 |
+
"learning_rate": 4.9724432121252836e-05,
|
| 4090 |
+
"loss": 24.9074,
|
| 4091 |
+
"step": 11660
|
| 4092 |
+
},
|
| 4093 |
+
{
|
| 4094 |
+
"epoch": 0.017301755654178196,
|
| 4095 |
+
"grad_norm": 7.375,
|
| 4096 |
+
"learning_rate": 4.972393822775207e-05,
|
| 4097 |
+
"loss": 24.8813,
|
| 4098 |
+
"step": 11680
|
| 4099 |
+
},
|
| 4100 |
+
{
|
| 4101 |
+
"epoch": 0.017331381948106583,
|
| 4102 |
+
"grad_norm": 7.40625,
|
| 4103 |
+
"learning_rate": 4.9723444334251316e-05,
|
| 4104 |
+
"loss": 24.7468,
|
| 4105 |
+
"step": 11700
|
| 4106 |
+
},
|
| 4107 |
+
{
|
| 4108 |
+
"epoch": 0.01736100824203497,
|
| 4109 |
+
"grad_norm": 7.65625,
|
| 4110 |
+
"learning_rate": 4.972295044075056e-05,
|
| 4111 |
+
"loss": 24.791,
|
| 4112 |
+
"step": 11720
|
| 4113 |
+
},
|
| 4114 |
+
{
|
| 4115 |
+
"epoch": 0.01739063453596336,
|
| 4116 |
+
"grad_norm": 7.3125,
|
| 4117 |
+
"learning_rate": 4.972245654724981e-05,
|
| 4118 |
+
"loss": 24.8677,
|
| 4119 |
+
"step": 11740
|
| 4120 |
+
},
|
| 4121 |
+
{
|
| 4122 |
+
"epoch": 0.017420260829891746,
|
| 4123 |
+
"grad_norm": 11.625,
|
| 4124 |
+
"learning_rate": 4.9721962653749047e-05,
|
| 4125 |
+
"loss": 24.8118,
|
| 4126 |
+
"step": 11760
|
| 4127 |
+
},
|
| 4128 |
+
{
|
| 4129 |
+
"epoch": 0.017449887123820132,
|
| 4130 |
+
"grad_norm": 7.15625,
|
| 4131 |
+
"learning_rate": 4.972146876024829e-05,
|
| 4132 |
+
"loss": 24.8483,
|
| 4133 |
+
"step": 11780
|
| 4134 |
+
},
|
| 4135 |
+
{
|
| 4136 |
+
"epoch": 0.017479513417748522,
|
| 4137 |
+
"grad_norm": 9.8125,
|
| 4138 |
+
"learning_rate": 4.972097486674754e-05,
|
| 4139 |
+
"loss": 24.8652,
|
| 4140 |
+
"step": 11800
|
| 4141 |
+
},
|
| 4142 |
+
{
|
| 4143 |
+
"epoch": 0.01750913971167691,
|
| 4144 |
+
"grad_norm": 7.4375,
|
| 4145 |
+
"learning_rate": 4.972048097324678e-05,
|
| 4146 |
+
"loss": 24.7769,
|
| 4147 |
+
"step": 11820
|
| 4148 |
+
},
|
| 4149 |
+
{
|
| 4150 |
+
"epoch": 0.017538766005605295,
|
| 4151 |
+
"grad_norm": 8.1875,
|
| 4152 |
+
"learning_rate": 4.971998707974602e-05,
|
| 4153 |
+
"loss": 24.8606,
|
| 4154 |
+
"step": 11840
|
| 4155 |
+
},
|
| 4156 |
+
{
|
| 4157 |
+
"epoch": 0.01756839229953368,
|
| 4158 |
+
"grad_norm": 8.25,
|
| 4159 |
+
"learning_rate": 4.9719493186245264e-05,
|
| 4160 |
+
"loss": 24.7523,
|
| 4161 |
+
"step": 11860
|
| 4162 |
+
},
|
| 4163 |
+
{
|
| 4164 |
+
"epoch": 0.01759801859346207,
|
| 4165 |
+
"grad_norm": 9.9375,
|
| 4166 |
+
"learning_rate": 4.9718999292744514e-05,
|
| 4167 |
+
"loss": 24.7848,
|
| 4168 |
+
"step": 11880
|
| 4169 |
+
},
|
| 4170 |
+
{
|
| 4171 |
+
"epoch": 0.017627644887390458,
|
| 4172 |
+
"grad_norm": 7.6875,
|
| 4173 |
+
"learning_rate": 4.971850539924375e-05,
|
| 4174 |
+
"loss": 24.7667,
|
| 4175 |
+
"step": 11900
|
| 4176 |
+
},
|
| 4177 |
+
{
|
| 4178 |
+
"epoch": 0.017657271181318844,
|
| 4179 |
+
"grad_norm": 9.0,
|
| 4180 |
+
"learning_rate": 4.9718011505742994e-05,
|
| 4181 |
+
"loss": 24.7369,
|
| 4182 |
+
"step": 11920
|
| 4183 |
+
},
|
| 4184 |
+
{
|
| 4185 |
+
"epoch": 0.01768689747524723,
|
| 4186 |
+
"grad_norm": 7.28125,
|
| 4187 |
+
"learning_rate": 4.971751761224224e-05,
|
| 4188 |
+
"loss": 24.7954,
|
| 4189 |
+
"step": 11940
|
| 4190 |
+
},
|
| 4191 |
+
{
|
| 4192 |
+
"epoch": 0.01771652376917562,
|
| 4193 |
+
"grad_norm": 8.4375,
|
| 4194 |
+
"learning_rate": 4.971702371874149e-05,
|
| 4195 |
+
"loss": 24.7734,
|
| 4196 |
+
"step": 11960
|
| 4197 |
+
},
|
| 4198 |
+
{
|
| 4199 |
+
"epoch": 0.017746150063104007,
|
| 4200 |
+
"grad_norm": 8.0,
|
| 4201 |
+
"learning_rate": 4.9716529825240724e-05,
|
| 4202 |
+
"loss": 24.8222,
|
| 4203 |
+
"step": 11980
|
| 4204 |
+
},
|
| 4205 |
+
{
|
| 4206 |
+
"epoch": 0.017775776357032393,
|
| 4207 |
+
"grad_norm": 10.625,
|
| 4208 |
+
"learning_rate": 4.971603593173997e-05,
|
| 4209 |
+
"loss": 24.7799,
|
| 4210 |
+
"step": 12000
|
| 4211 |
}
|
| 4212 |
],
|
| 4213 |
"logging_steps": 20,
|
|
|
|
| 4227 |
"attributes": {}
|
| 4228 |
}
|
| 4229 |
},
|
| 4230 |
+
"total_flos": 2.22031198508611e+19,
|
| 4231 |
"train_batch_size": 48,
|
| 4232 |
"trial_name": null,
|
| 4233 |
"trial_params": null
|