Training in progress, step 15000, 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 487156538
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a6597692baaaecd56cc3119530c69265f378d69e034a383c06ed9343d53fe0ea
|
| 3 |
size 487156538
|
last-checkpoint/pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1059459406
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:75301060d0614bae90398039500c97ff95194395b69fb1d587d503e800901639
|
| 3 |
size 1059459406
|
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 sha256:58725715e3a81fd995f3eb6fcd80daea9b56d116fca9b35a744c99f27b82bcc1
|
| 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 sha256:9a83fee6b01b58aa2d91d3e7341c75d9a4ecceb467333bab32df1edbfad5b705
|
| 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:ba3120a58a84b571dd9da9df7ca01c8edbdf1e3273712bd70e4eea91eb7a2a07
|
| 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:f845f06854dde41c169725cc192fbd9612a41feb786e9ebb38d8ffe1fdb6a40a
|
| 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 sha256:1f1eef33864701acffbe1e463e988c61c385ad81d27a75b2f5773dcc59db84b1
|
| 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,
|
|
@@ -4908,6 +4908,356 @@
|
|
| 4908 |
"learning_rate": 0.0004966664658166431,
|
| 4909 |
"loss": 18.6651,
|
| 4910 |
"step": 14000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4911 |
}
|
| 4912 |
],
|
| 4913 |
"logging_steps": 20,
|
|
@@ -4927,7 +5277,7 @@
|
|
| 4927 |
"attributes": {}
|
| 4928 |
}
|
| 4929 |
},
|
| 4930 |
-
"total_flos": 3.
|
| 4931 |
"train_batch_size": 48,
|
| 4932 |
"trial_name": null,
|
| 4933 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.022219720446290493,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 15000,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 4908 |
"learning_rate": 0.0004966664658166431,
|
| 4909 |
"loss": 18.6651,
|
| 4910 |
"step": 14000
|
| 4911 |
+
},
|
| 4912 |
+
{
|
| 4913 |
+
"epoch": 0.02076803204379951,
|
| 4914 |
+
"grad_norm": 8.6875,
|
| 4915 |
+
"learning_rate": 0.0004966615268816355,
|
| 4916 |
+
"loss": 18.6887,
|
| 4917 |
+
"step": 14020
|
| 4918 |
+
},
|
| 4919 |
+
{
|
| 4920 |
+
"epoch": 0.0207976583377279,
|
| 4921 |
+
"grad_norm": 8.5,
|
| 4922 |
+
"learning_rate": 0.0004966565879466278,
|
| 4923 |
+
"loss": 18.6563,
|
| 4924 |
+
"step": 14040
|
| 4925 |
+
},
|
| 4926 |
+
{
|
| 4927 |
+
"epoch": 0.020827284631656288,
|
| 4928 |
+
"grad_norm": 7.4375,
|
| 4929 |
+
"learning_rate": 0.0004966516490116204,
|
| 4930 |
+
"loss": 18.6696,
|
| 4931 |
+
"step": 14060
|
| 4932 |
+
},
|
| 4933 |
+
{
|
| 4934 |
+
"epoch": 0.020856910925584674,
|
| 4935 |
+
"grad_norm": 13.0,
|
| 4936 |
+
"learning_rate": 0.0004966467100766127,
|
| 4937 |
+
"loss": 18.7075,
|
| 4938 |
+
"step": 14080
|
| 4939 |
+
},
|
| 4940 |
+
{
|
| 4941 |
+
"epoch": 0.02088653721951306,
|
| 4942 |
+
"grad_norm": 8.25,
|
| 4943 |
+
"learning_rate": 0.0004966417711416051,
|
| 4944 |
+
"loss": 18.7774,
|
| 4945 |
+
"step": 14100
|
| 4946 |
+
},
|
| 4947 |
+
{
|
| 4948 |
+
"epoch": 0.02091616351344145,
|
| 4949 |
+
"grad_norm": 8.0625,
|
| 4950 |
+
"learning_rate": 0.0004966368322065976,
|
| 4951 |
+
"loss": 18.7288,
|
| 4952 |
+
"step": 14120
|
| 4953 |
+
},
|
| 4954 |
+
{
|
| 4955 |
+
"epoch": 0.020945789807369837,
|
| 4956 |
+
"grad_norm": 6.9375,
|
| 4957 |
+
"learning_rate": 0.00049663189327159,
|
| 4958 |
+
"loss": 18.7008,
|
| 4959 |
+
"step": 14140
|
| 4960 |
+
},
|
| 4961 |
+
{
|
| 4962 |
+
"epoch": 0.020975416101298223,
|
| 4963 |
+
"grad_norm": 8.25,
|
| 4964 |
+
"learning_rate": 0.0004966269543365825,
|
| 4965 |
+
"loss": 18.6659,
|
| 4966 |
+
"step": 14160
|
| 4967 |
+
},
|
| 4968 |
+
{
|
| 4969 |
+
"epoch": 0.02100504239522661,
|
| 4970 |
+
"grad_norm": 7.625,
|
| 4971 |
+
"learning_rate": 0.0004966220154015749,
|
| 4972 |
+
"loss": 18.5981,
|
| 4973 |
+
"step": 14180
|
| 4974 |
+
},
|
| 4975 |
+
{
|
| 4976 |
+
"epoch": 0.021034668689155,
|
| 4977 |
+
"grad_norm": 7.53125,
|
| 4978 |
+
"learning_rate": 0.0004966170764665674,
|
| 4979 |
+
"loss": 18.6925,
|
| 4980 |
+
"step": 14200
|
| 4981 |
+
},
|
| 4982 |
+
{
|
| 4983 |
+
"epoch": 0.021064294983083386,
|
| 4984 |
+
"grad_norm": 8.125,
|
| 4985 |
+
"learning_rate": 0.0004966121375315598,
|
| 4986 |
+
"loss": 18.7061,
|
| 4987 |
+
"step": 14220
|
| 4988 |
+
},
|
| 4989 |
+
{
|
| 4990 |
+
"epoch": 0.021093921277011773,
|
| 4991 |
+
"grad_norm": 10.3125,
|
| 4992 |
+
"learning_rate": 0.0004966071985965522,
|
| 4993 |
+
"loss": 18.6827,
|
| 4994 |
+
"step": 14240
|
| 4995 |
+
},
|
| 4996 |
+
{
|
| 4997 |
+
"epoch": 0.02112354757094016,
|
| 4998 |
+
"grad_norm": 7.875,
|
| 4999 |
+
"learning_rate": 0.0004966022596615446,
|
| 5000 |
+
"loss": 18.6601,
|
| 5001 |
+
"step": 14260
|
| 5002 |
+
},
|
| 5003 |
+
{
|
| 5004 |
+
"epoch": 0.02115317386486855,
|
| 5005 |
+
"grad_norm": 9.5625,
|
| 5006 |
+
"learning_rate": 0.0004965973207265371,
|
| 5007 |
+
"loss": 18.6155,
|
| 5008 |
+
"step": 14280
|
| 5009 |
+
},
|
| 5010 |
+
{
|
| 5011 |
+
"epoch": 0.021182800158796936,
|
| 5012 |
+
"grad_norm": 9.375,
|
| 5013 |
+
"learning_rate": 0.0004965923817915295,
|
| 5014 |
+
"loss": 18.6963,
|
| 5015 |
+
"step": 14300
|
| 5016 |
+
},
|
| 5017 |
+
{
|
| 5018 |
+
"epoch": 0.021212426452725322,
|
| 5019 |
+
"grad_norm": 7.34375,
|
| 5020 |
+
"learning_rate": 0.000496587442856522,
|
| 5021 |
+
"loss": 18.5988,
|
| 5022 |
+
"step": 14320
|
| 5023 |
+
},
|
| 5024 |
+
{
|
| 5025 |
+
"epoch": 0.02124205274665371,
|
| 5026 |
+
"grad_norm": 7.78125,
|
| 5027 |
+
"learning_rate": 0.0004965825039215144,
|
| 5028 |
+
"loss": 18.6373,
|
| 5029 |
+
"step": 14340
|
| 5030 |
+
},
|
| 5031 |
+
{
|
| 5032 |
+
"epoch": 0.0212716790405821,
|
| 5033 |
+
"grad_norm": 8.125,
|
| 5034 |
+
"learning_rate": 0.0004965775649865069,
|
| 5035 |
+
"loss": 18.6575,
|
| 5036 |
+
"step": 14360
|
| 5037 |
+
},
|
| 5038 |
+
{
|
| 5039 |
+
"epoch": 0.021301305334510485,
|
| 5040 |
+
"grad_norm": 13.1875,
|
| 5041 |
+
"learning_rate": 0.0004965726260514993,
|
| 5042 |
+
"loss": 18.6643,
|
| 5043 |
+
"step": 14380
|
| 5044 |
+
},
|
| 5045 |
+
{
|
| 5046 |
+
"epoch": 0.02133093162843887,
|
| 5047 |
+
"grad_norm": 8.25,
|
| 5048 |
+
"learning_rate": 0.0004965676871164917,
|
| 5049 |
+
"loss": 18.7713,
|
| 5050 |
+
"step": 14400
|
| 5051 |
+
},
|
| 5052 |
+
{
|
| 5053 |
+
"epoch": 0.021360557922367258,
|
| 5054 |
+
"grad_norm": 7.8125,
|
| 5055 |
+
"learning_rate": 0.0004965627481814841,
|
| 5056 |
+
"loss": 18.66,
|
| 5057 |
+
"step": 14420
|
| 5058 |
+
},
|
| 5059 |
+
{
|
| 5060 |
+
"epoch": 0.021390184216295648,
|
| 5061 |
+
"grad_norm": 7.40625,
|
| 5062 |
+
"learning_rate": 0.0004965578092464766,
|
| 5063 |
+
"loss": 18.6699,
|
| 5064 |
+
"step": 14440
|
| 5065 |
+
},
|
| 5066 |
+
{
|
| 5067 |
+
"epoch": 0.021419810510224034,
|
| 5068 |
+
"grad_norm": 8.5625,
|
| 5069 |
+
"learning_rate": 0.000496552870311469,
|
| 5070 |
+
"loss": 18.617,
|
| 5071 |
+
"step": 14460
|
| 5072 |
+
},
|
| 5073 |
+
{
|
| 5074 |
+
"epoch": 0.02144943680415242,
|
| 5075 |
+
"grad_norm": 8.1875,
|
| 5076 |
+
"learning_rate": 0.0004965479313764614,
|
| 5077 |
+
"loss": 18.6727,
|
| 5078 |
+
"step": 14480
|
| 5079 |
+
},
|
| 5080 |
+
{
|
| 5081 |
+
"epoch": 0.021479063098080807,
|
| 5082 |
+
"grad_norm": 7.5625,
|
| 5083 |
+
"learning_rate": 0.0004965429924414539,
|
| 5084 |
+
"loss": 18.5574,
|
| 5085 |
+
"step": 14500
|
| 5086 |
+
},
|
| 5087 |
+
{
|
| 5088 |
+
"epoch": 0.021508689392009197,
|
| 5089 |
+
"grad_norm": 8.875,
|
| 5090 |
+
"learning_rate": 0.0004965380535064463,
|
| 5091 |
+
"loss": 18.7566,
|
| 5092 |
+
"step": 14520
|
| 5093 |
+
},
|
| 5094 |
+
{
|
| 5095 |
+
"epoch": 0.021538315685937583,
|
| 5096 |
+
"grad_norm": 8.3125,
|
| 5097 |
+
"learning_rate": 0.0004965331145714388,
|
| 5098 |
+
"loss": 18.7108,
|
| 5099 |
+
"step": 14540
|
| 5100 |
+
},
|
| 5101 |
+
{
|
| 5102 |
+
"epoch": 0.02156794197986597,
|
| 5103 |
+
"grad_norm": 7.09375,
|
| 5104 |
+
"learning_rate": 0.0004965281756364312,
|
| 5105 |
+
"loss": 18.5922,
|
| 5106 |
+
"step": 14560
|
| 5107 |
+
},
|
| 5108 |
+
{
|
| 5109 |
+
"epoch": 0.021597568273794356,
|
| 5110 |
+
"grad_norm": 7.125,
|
| 5111 |
+
"learning_rate": 0.0004965232367014237,
|
| 5112 |
+
"loss": 18.6615,
|
| 5113 |
+
"step": 14580
|
| 5114 |
+
},
|
| 5115 |
+
{
|
| 5116 |
+
"epoch": 0.021627194567722746,
|
| 5117 |
+
"grad_norm": 7.5625,
|
| 5118 |
+
"learning_rate": 0.0004965182977664161,
|
| 5119 |
+
"loss": 18.6649,
|
| 5120 |
+
"step": 14600
|
| 5121 |
+
},
|
| 5122 |
+
{
|
| 5123 |
+
"epoch": 0.021656820861651133,
|
| 5124 |
+
"grad_norm": 8.3125,
|
| 5125 |
+
"learning_rate": 0.0004965133588314085,
|
| 5126 |
+
"loss": 18.7153,
|
| 5127 |
+
"step": 14620
|
| 5128 |
+
},
|
| 5129 |
+
{
|
| 5130 |
+
"epoch": 0.02168644715557952,
|
| 5131 |
+
"grad_norm": 7.6875,
|
| 5132 |
+
"learning_rate": 0.0004965084198964008,
|
| 5133 |
+
"loss": 18.6446,
|
| 5134 |
+
"step": 14640
|
| 5135 |
+
},
|
| 5136 |
+
{
|
| 5137 |
+
"epoch": 0.021716073449507906,
|
| 5138 |
+
"grad_norm": 6.90625,
|
| 5139 |
+
"learning_rate": 0.0004965034809613934,
|
| 5140 |
+
"loss": 18.6951,
|
| 5141 |
+
"step": 14660
|
| 5142 |
+
},
|
| 5143 |
+
{
|
| 5144 |
+
"epoch": 0.021745699743436295,
|
| 5145 |
+
"grad_norm": 7.875,
|
| 5146 |
+
"learning_rate": 0.0004964985420263857,
|
| 5147 |
+
"loss": 18.6594,
|
| 5148 |
+
"step": 14680
|
| 5149 |
+
},
|
| 5150 |
+
{
|
| 5151 |
+
"epoch": 0.021775326037364682,
|
| 5152 |
+
"grad_norm": 9.25,
|
| 5153 |
+
"learning_rate": 0.0004964936030913782,
|
| 5154 |
+
"loss": 18.6275,
|
| 5155 |
+
"step": 14700
|
| 5156 |
+
},
|
| 5157 |
+
{
|
| 5158 |
+
"epoch": 0.02180495233129307,
|
| 5159 |
+
"grad_norm": 16.5,
|
| 5160 |
+
"learning_rate": 0.0004964886641563706,
|
| 5161 |
+
"loss": 18.6842,
|
| 5162 |
+
"step": 14720
|
| 5163 |
+
},
|
| 5164 |
+
{
|
| 5165 |
+
"epoch": 0.021834578625221458,
|
| 5166 |
+
"grad_norm": 8.5625,
|
| 5167 |
+
"learning_rate": 0.0004964837252213631,
|
| 5168 |
+
"loss": 18.5828,
|
| 5169 |
+
"step": 14740
|
| 5170 |
+
},
|
| 5171 |
+
{
|
| 5172 |
+
"epoch": 0.021864204919149845,
|
| 5173 |
+
"grad_norm": 8.0,
|
| 5174 |
+
"learning_rate": 0.0004964787862863555,
|
| 5175 |
+
"loss": 18.5667,
|
| 5176 |
+
"step": 14760
|
| 5177 |
+
},
|
| 5178 |
+
{
|
| 5179 |
+
"epoch": 0.02189383121307823,
|
| 5180 |
+
"grad_norm": 7.8125,
|
| 5181 |
+
"learning_rate": 0.0004964738473513479,
|
| 5182 |
+
"loss": 18.575,
|
| 5183 |
+
"step": 14780
|
| 5184 |
+
},
|
| 5185 |
+
{
|
| 5186 |
+
"epoch": 0.021923457507006618,
|
| 5187 |
+
"grad_norm": 14.8125,
|
| 5188 |
+
"learning_rate": 0.0004964689084163404,
|
| 5189 |
+
"loss": 18.5941,
|
| 5190 |
+
"step": 14800
|
| 5191 |
+
},
|
| 5192 |
+
{
|
| 5193 |
+
"epoch": 0.021953083800935008,
|
| 5194 |
+
"grad_norm": 7.875,
|
| 5195 |
+
"learning_rate": 0.0004964639694813328,
|
| 5196 |
+
"loss": 18.5917,
|
| 5197 |
+
"step": 14820
|
| 5198 |
+
},
|
| 5199 |
+
{
|
| 5200 |
+
"epoch": 0.021982710094863394,
|
| 5201 |
+
"grad_norm": 9.75,
|
| 5202 |
+
"learning_rate": 0.0004964590305463252,
|
| 5203 |
+
"loss": 18.6876,
|
| 5204 |
+
"step": 14840
|
| 5205 |
+
},
|
| 5206 |
+
{
|
| 5207 |
+
"epoch": 0.02201233638879178,
|
| 5208 |
+
"grad_norm": 6.8125,
|
| 5209 |
+
"learning_rate": 0.0004964540916113176,
|
| 5210 |
+
"loss": 18.5439,
|
| 5211 |
+
"step": 14860
|
| 5212 |
+
},
|
| 5213 |
+
{
|
| 5214 |
+
"epoch": 0.022041962682720167,
|
| 5215 |
+
"grad_norm": 7.96875,
|
| 5216 |
+
"learning_rate": 0.0004964491526763101,
|
| 5217 |
+
"loss": 18.5891,
|
| 5218 |
+
"step": 14880
|
| 5219 |
+
},
|
| 5220 |
+
{
|
| 5221 |
+
"epoch": 0.022071588976648557,
|
| 5222 |
+
"grad_norm": 7.4375,
|
| 5223 |
+
"learning_rate": 0.0004964442137413025,
|
| 5224 |
+
"loss": 18.585,
|
| 5225 |
+
"step": 14900
|
| 5226 |
+
},
|
| 5227 |
+
{
|
| 5228 |
+
"epoch": 0.022101215270576943,
|
| 5229 |
+
"grad_norm": 8.3125,
|
| 5230 |
+
"learning_rate": 0.000496439274806295,
|
| 5231 |
+
"loss": 18.6264,
|
| 5232 |
+
"step": 14920
|
| 5233 |
+
},
|
| 5234 |
+
{
|
| 5235 |
+
"epoch": 0.02213084156450533,
|
| 5236 |
+
"grad_norm": 8.125,
|
| 5237 |
+
"learning_rate": 0.0004964343358712874,
|
| 5238 |
+
"loss": 18.6338,
|
| 5239 |
+
"step": 14940
|
| 5240 |
+
},
|
| 5241 |
+
{
|
| 5242 |
+
"epoch": 0.022160467858433716,
|
| 5243 |
+
"grad_norm": 8.1875,
|
| 5244 |
+
"learning_rate": 0.0004964293969362799,
|
| 5245 |
+
"loss": 18.5912,
|
| 5246 |
+
"step": 14960
|
| 5247 |
+
},
|
| 5248 |
+
{
|
| 5249 |
+
"epoch": 0.022190094152362106,
|
| 5250 |
+
"grad_norm": 7.6875,
|
| 5251 |
+
"learning_rate": 0.0004964244580012723,
|
| 5252 |
+
"loss": 18.6101,
|
| 5253 |
+
"step": 14980
|
| 5254 |
+
},
|
| 5255 |
+
{
|
| 5256 |
+
"epoch": 0.022219720446290493,
|
| 5257 |
+
"grad_norm": 7.59375,
|
| 5258 |
+
"learning_rate": 0.0004964195190662647,
|
| 5259 |
+
"loss": 18.5823,
|
| 5260 |
+
"step": 15000
|
| 5261 |
}
|
| 5262 |
],
|
| 5263 |
"logging_steps": 20,
|
|
|
|
| 5277 |
"attributes": {}
|
| 5278 |
}
|
| 5279 |
},
|
| 5280 |
+
"total_flos": 3.33834741124327e+19,
|
| 5281 |
"train_batch_size": 48,
|
| 5282 |
"trial_name": null,
|
| 5283 |
"trial_params": null
|