Training in progress, epoch 3, checkpoint
Browse files- last-checkpoint/model-00001-of-00004.safetensors +1 -1
- last-checkpoint/model-00002-of-00004.safetensors +1 -1
- last-checkpoint/model-00003-of-00004.safetensors +1 -1
- last-checkpoint/model-00004-of-00004.safetensors +1 -1
- last-checkpoint/optimizer.pt +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/rng_state_4.pth +1 -1
- last-checkpoint/rng_state_5.pth +1 -1
- last-checkpoint/rng_state_6.pth +1 -1
- last-checkpoint/rng_state_7.pth +1 -1
- last-checkpoint/scheduler.pt +1 -1
- last-checkpoint/trainer_state.json +516 -4
last-checkpoint/model-00001-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4976698672
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9d3b18816811918972cba42da78818e3c7aa5d8e31f822375470c2f40641e8a2
|
| 3 |
size 4976698672
|
last-checkpoint/model-00002-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4999802720
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:609faa192d9a31b460e1e9d369f7a2bb92f9ecbedb8d282dfd9a9d86e77b82e2
|
| 3 |
size 4999802720
|
last-checkpoint/model-00003-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4915916176
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1187480a882d25ada76753b8c72449b0e409556587c1ef893ce5b508bb4bb2ba
|
| 3 |
size 4915916176
|
last-checkpoint/model-00004-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1168138808
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e7c05715f857592efaf0022aee229538b0cbd4bc4b8784e7e8d58053ec6d017e
|
| 3 |
size 1168138808
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 32121299754
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e3d3fec74b231513b476c27b7eda7f7d9835dec42b905326e0577ccee3a0cc3
|
| 3 |
size 32121299754
|
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 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:69ec6e3926fa071bede113523efa3dc6e630c3c7958c54a9ca321cf4d62ed145
|
| 3 |
size 15984
|
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 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f6127ee4f0c13500ec5038fce65af8f7beec63c137c7d4b7c157aa6303cf5879
|
| 3 |
size 15984
|
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 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:da01d1c5eb2cc3a323f97c1f590d13ccfac2a4c5b1479bd378b4e643304f5a4f
|
| 3 |
size 15984
|
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 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:49a3f04d76c0d3acc7d3dd95a04215f368f35a451ae8cba8a2fdba38cda9ca0a
|
| 3 |
size 15984
|
last-checkpoint/rng_state_4.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:df7d2c9825dba80cb544920f8cc0c72122f96514e6cd259052a8765b034393e2
|
| 3 |
size 15984
|
last-checkpoint/rng_state_5.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a20a42d44ff48cc162224010190e898fe28598ddad8cd1896d330a3bb1d8ec3
|
| 3 |
size 15984
|
last-checkpoint/rng_state_6.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:18ac0dc4f09f25179860561fcea7c5c8f997aabdc46a170665f9dc5a72bc27c6
|
| 3 |
size 15984
|
last-checkpoint/rng_state_7.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a16fcb5411ff961b47eff7378d85105fe9837e0492d19ea5ce3b7c4b77aa3b6
|
| 3 |
size 15984
|
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 sha256:04a46754a7468c445356d55f12e6f57375db57b7d43b8c6963579dc82de997b3
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 300,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -1045,6 +1045,518 @@
|
|
| 1045 |
"learning_rate": 5.041773562018135e-07,
|
| 1046 |
"loss": 1.7019,
|
| 1047 |
"step": 725
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1048 |
}
|
| 1049 |
],
|
| 1050 |
"logging_steps": 5,
|
|
@@ -1059,12 +1571,12 @@
|
|
| 1059 |
"should_evaluate": false,
|
| 1060 |
"should_log": false,
|
| 1061 |
"should_save": true,
|
| 1062 |
-
"should_training_stop":
|
| 1063 |
},
|
| 1064 |
"attributes": {}
|
| 1065 |
}
|
| 1066 |
},
|
| 1067 |
-
"total_flos":
|
| 1068 |
"train_batch_size": 8,
|
| 1069 |
"trial_name": null,
|
| 1070 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
"eval_steps": 300,
|
| 6 |
+
"global_step": 1089,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1045 |
"learning_rate": 5.041773562018135e-07,
|
| 1046 |
"loss": 1.7019,
|
| 1047 |
"step": 725
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 2.0110192837465566,
|
| 1051 |
+
"grad_norm": 1.265625,
|
| 1052 |
+
"learning_rate": 4.91680188206047e-07,
|
| 1053 |
+
"loss": 1.7011,
|
| 1054 |
+
"step": 730
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 2.024793388429752,
|
| 1058 |
+
"grad_norm": 1.265625,
|
| 1059 |
+
"learning_rate": 4.792891678019115e-07,
|
| 1060 |
+
"loss": 1.7013,
|
| 1061 |
+
"step": 735
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 2.038567493112948,
|
| 1065 |
+
"grad_norm": 1.2734375,
|
| 1066 |
+
"learning_rate": 4.6700688248834664e-07,
|
| 1067 |
+
"loss": 1.6895,
|
| 1068 |
+
"step": 740
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 2.0523415977961434,
|
| 1072 |
+
"grad_norm": 1.3125,
|
| 1073 |
+
"learning_rate": 4.548358970581757e-07,
|
| 1074 |
+
"loss": 1.7029,
|
| 1075 |
+
"step": 745
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 2.0661157024793386,
|
| 1079 |
+
"grad_norm": 1.2734375,
|
| 1080 |
+
"learning_rate": 4.427787530625278e-07,
|
| 1081 |
+
"loss": 1.6931,
|
| 1082 |
+
"step": 750
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 2.0798898071625342,
|
| 1086 |
+
"grad_norm": 1.2734375,
|
| 1087 |
+
"learning_rate": 4.3083796828010675e-07,
|
| 1088 |
+
"loss": 1.6886,
|
| 1089 |
+
"step": 755
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 2.09366391184573,
|
| 1093 |
+
"grad_norm": 1.3046875,
|
| 1094 |
+
"learning_rate": 4.190160361914292e-07,
|
| 1095 |
+
"loss": 1.6907,
|
| 1096 |
+
"step": 760
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 2.1074380165289255,
|
| 1100 |
+
"grad_norm": 1.296875,
|
| 1101 |
+
"learning_rate": 4.07315425458134e-07,
|
| 1102 |
+
"loss": 1.6924,
|
| 1103 |
+
"step": 765
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 2.121212121212121,
|
| 1107 |
+
"grad_norm": 1.28125,
|
| 1108 |
+
"learning_rate": 3.9573857940747537e-07,
|
| 1109 |
+
"loss": 1.7019,
|
| 1110 |
+
"step": 770
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 2.1349862258953167,
|
| 1114 |
+
"grad_norm": 1.28125,
|
| 1115 |
+
"learning_rate": 3.8428791552210594e-07,
|
| 1116 |
+
"loss": 1.6975,
|
| 1117 |
+
"step": 775
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 2.1487603305785123,
|
| 1121 |
+
"grad_norm": 1.2890625,
|
| 1122 |
+
"learning_rate": 3.729658249352563e-07,
|
| 1123 |
+
"loss": 1.6986,
|
| 1124 |
+
"step": 780
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 2.162534435261708,
|
| 1128 |
+
"grad_norm": 1.3125,
|
| 1129 |
+
"learning_rate": 3.6177467193141886e-07,
|
| 1130 |
+
"loss": 1.6893,
|
| 1131 |
+
"step": 785
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 2.1763085399449036,
|
| 1135 |
+
"grad_norm": 1.2734375,
|
| 1136 |
+
"learning_rate": 3.5071679345263537e-07,
|
| 1137 |
+
"loss": 1.6833,
|
| 1138 |
+
"step": 790
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 2.190082644628099,
|
| 1142 |
+
"grad_norm": 1.2734375,
|
| 1143 |
+
"learning_rate": 3.397944986104968e-07,
|
| 1144 |
+
"loss": 1.693,
|
| 1145 |
+
"step": 795
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 2.203856749311295,
|
| 1149 |
+
"grad_norm": 1.3046875,
|
| 1150 |
+
"learning_rate": 3.290100682039516e-07,
|
| 1151 |
+
"loss": 1.6978,
|
| 1152 |
+
"step": 800
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 2.2176308539944904,
|
| 1156 |
+
"grad_norm": 1.28125,
|
| 1157 |
+
"learning_rate": 3.1836575424303034e-07,
|
| 1158 |
+
"loss": 1.7019,
|
| 1159 |
+
"step": 805
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 2.231404958677686,
|
| 1163 |
+
"grad_norm": 1.2890625,
|
| 1164 |
+
"learning_rate": 3.078637794785791e-07,
|
| 1165 |
+
"loss": 1.6977,
|
| 1166 |
+
"step": 810
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 2.2451790633608817,
|
| 1170 |
+
"grad_norm": 1.28125,
|
| 1171 |
+
"learning_rate": 2.9750633693810224e-07,
|
| 1172 |
+
"loss": 1.6898,
|
| 1173 |
+
"step": 815
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 2.2589531680440773,
|
| 1177 |
+
"grad_norm": 1.28125,
|
| 1178 |
+
"learning_rate": 2.872955894678153e-07,
|
| 1179 |
+
"loss": 1.6915,
|
| 1180 |
+
"step": 820
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 2.2727272727272725,
|
| 1184 |
+
"grad_norm": 1.265625,
|
| 1185 |
+
"learning_rate": 2.7723366928099754e-07,
|
| 1186 |
+
"loss": 1.6922,
|
| 1187 |
+
"step": 825
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 2.2865013774104685,
|
| 1191 |
+
"grad_norm": 1.3046875,
|
| 1192 |
+
"learning_rate": 2.673226775127422e-07,
|
| 1193 |
+
"loss": 1.6922,
|
| 1194 |
+
"step": 830
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 2.3002754820936637,
|
| 1198 |
+
"grad_norm": 1.3046875,
|
| 1199 |
+
"learning_rate": 2.5756468378119533e-07,
|
| 1200 |
+
"loss": 1.6873,
|
| 1201 |
+
"step": 835
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 2.3140495867768593,
|
| 1205 |
+
"grad_norm": 1.265625,
|
| 1206 |
+
"learning_rate": 2.4796172575537934e-07,
|
| 1207 |
+
"loss": 1.7068,
|
| 1208 |
+
"step": 840
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 2.327823691460055,
|
| 1212 |
+
"grad_norm": 1.3203125,
|
| 1213 |
+
"learning_rate": 2.3851580872968435e-07,
|
| 1214 |
+
"loss": 1.6993,
|
| 1215 |
+
"step": 845
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 2.3415977961432506,
|
| 1219 |
+
"grad_norm": 1.3046875,
|
| 1220 |
+
"learning_rate": 2.292289052051224e-07,
|
| 1221 |
+
"loss": 1.6992,
|
| 1222 |
+
"step": 850
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 2.355371900826446,
|
| 1226 |
+
"grad_norm": 1.2578125,
|
| 1227 |
+
"learning_rate": 2.2010295447742743e-07,
|
| 1228 |
+
"loss": 1.6891,
|
| 1229 |
+
"step": 855
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 2.369146005509642,
|
| 1233 |
+
"grad_norm": 1.296875,
|
| 1234 |
+
"learning_rate": 2.111398622320927e-07,
|
| 1235 |
+
"loss": 1.6968,
|
| 1236 |
+
"step": 860
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 2.3829201101928374,
|
| 1240 |
+
"grad_norm": 1.296875,
|
| 1241 |
+
"learning_rate": 2.0234150014642305e-07,
|
| 1242 |
+
"loss": 1.6946,
|
| 1243 |
+
"step": 865
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 2.396694214876033,
|
| 1247 |
+
"grad_norm": 1.265625,
|
| 1248 |
+
"learning_rate": 1.937097054986915e-07,
|
| 1249 |
+
"loss": 1.6892,
|
| 1250 |
+
"step": 870
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 2.4104683195592287,
|
| 1254 |
+
"grad_norm": 1.2578125,
|
| 1255 |
+
"learning_rate": 1.8524628078447602e-07,
|
| 1256 |
+
"loss": 1.6915,
|
| 1257 |
+
"step": 875
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 2.4242424242424243,
|
| 1261 |
+
"grad_norm": 1.2890625,
|
| 1262 |
+
"learning_rate": 1.769529933402637e-07,
|
| 1263 |
+
"loss": 1.6946,
|
| 1264 |
+
"step": 880
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 2.43801652892562,
|
| 1268 |
+
"grad_norm": 1.2578125,
|
| 1269 |
+
"learning_rate": 1.6883157497439349e-07,
|
| 1270 |
+
"loss": 1.6975,
|
| 1271 |
+
"step": 885
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 2.4517906336088156,
|
| 1275 |
+
"grad_norm": 1.2734375,
|
| 1276 |
+
"learning_rate": 1.6088372160541962e-07,
|
| 1277 |
+
"loss": 1.6871,
|
| 1278 |
+
"step": 890
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 2.465564738292011,
|
| 1282 |
+
"grad_norm": 1.2734375,
|
| 1283 |
+
"learning_rate": 1.531110929079681e-07,
|
| 1284 |
+
"loss": 1.6909,
|
| 1285 |
+
"step": 895
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 2.479338842975207,
|
| 1289 |
+
"grad_norm": 1.28125,
|
| 1290 |
+
"learning_rate": 1.4551531196616396e-07,
|
| 1291 |
+
"loss": 1.6908,
|
| 1292 |
+
"step": 900
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 2.479338842975207,
|
| 1296 |
+
"eval_loss": 1.7025996446609497,
|
| 1297 |
+
"eval_runtime": 8.3873,
|
| 1298 |
+
"eval_samples_per_second": 83.936,
|
| 1299 |
+
"eval_steps_per_second": 2.623,
|
| 1300 |
+
"step": 900
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 2.4931129476584024,
|
| 1304 |
+
"grad_norm": 1.265625,
|
| 1305 |
+
"learning_rate": 1.3809796493469728e-07,
|
| 1306 |
+
"loss": 1.6981,
|
| 1307 |
+
"step": 905
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 2.5068870523415976,
|
| 1311 |
+
"grad_norm": 1.3046875,
|
| 1312 |
+
"learning_rate": 1.3086060070760196e-07,
|
| 1313 |
+
"loss": 1.6902,
|
| 1314 |
+
"step": 910
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 2.5206611570247937,
|
| 1318 |
+
"grad_norm": 1.3046875,
|
| 1319 |
+
"learning_rate": 1.23804730594814e-07,
|
| 1320 |
+
"loss": 1.6964,
|
| 1321 |
+
"step": 915
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 2.534435261707989,
|
| 1325 |
+
"grad_norm": 1.2890625,
|
| 1326 |
+
"learning_rate": 1.1693182800658042e-07,
|
| 1327 |
+
"loss": 1.6884,
|
| 1328 |
+
"step": 920
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 2.5482093663911844,
|
| 1332 |
+
"grad_norm": 1.2890625,
|
| 1333 |
+
"learning_rate": 1.102433281457802e-07,
|
| 1334 |
+
"loss": 1.6969,
|
| 1335 |
+
"step": 925
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 2.56198347107438,
|
| 1339 |
+
"grad_norm": 1.265625,
|
| 1340 |
+
"learning_rate": 1.0374062770822411e-07,
|
| 1341 |
+
"loss": 1.7003,
|
| 1342 |
+
"step": 930
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 2.5757575757575757,
|
| 1346 |
+
"grad_norm": 1.2578125,
|
| 1347 |
+
"learning_rate": 9.742508459099707e-08,
|
| 1348 |
+
"loss": 1.7095,
|
| 1349 |
+
"step": 935
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 2.5895316804407713,
|
| 1353 |
+
"grad_norm": 1.2890625,
|
| 1354 |
+
"learning_rate": 9.129801760890076e-08,
|
| 1355 |
+
"loss": 1.7026,
|
| 1356 |
+
"step": 940
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 2.603305785123967,
|
| 1360 |
+
"grad_norm": 1.2578125,
|
| 1361 |
+
"learning_rate": 8.536070621905811e-08,
|
| 1362 |
+
"loss": 1.6964,
|
| 1363 |
+
"step": 945
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 2.6170798898071626,
|
| 1367 |
+
"grad_norm": 1.3046875,
|
| 1368 |
+
"learning_rate": 7.961439025373617e-08,
|
| 1369 |
+
"loss": 1.6984,
|
| 1370 |
+
"step": 950
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 2.630853994490358,
|
| 1374 |
+
"grad_norm": 1.28125,
|
| 1375 |
+
"learning_rate": 7.40602696614444e-08,
|
| 1376 |
+
"loss": 1.7022,
|
| 1377 |
+
"step": 955
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 2.644628099173554,
|
| 1381 |
+
"grad_norm": 1.2734375,
|
| 1382 |
+
"learning_rate": 6.869950425636095e-08,
|
| 1383 |
+
"loss": 1.6955,
|
| 1384 |
+
"step": 960
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 2.6584022038567494,
|
| 1388 |
+
"grad_norm": 1.265625,
|
| 1389 |
+
"learning_rate": 6.353321347613815e-08,
|
| 1390 |
+
"loss": 1.6962,
|
| 1391 |
+
"step": 965
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 2.672176308539945,
|
| 1395 |
+
"grad_norm": 1.3828125,
|
| 1396 |
+
"learning_rate": 5.856247614814292e-08,
|
| 1397 |
+
"loss": 1.6914,
|
| 1398 |
+
"step": 970
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"epoch": 2.6859504132231407,
|
| 1402 |
+
"grad_norm": 1.296875,
|
| 1403 |
+
"learning_rate": 5.3788330264174506e-08,
|
| 1404 |
+
"loss": 1.6934,
|
| 1405 |
+
"step": 975
|
| 1406 |
+
},
|
| 1407 |
+
{
|
| 1408 |
+
"epoch": 2.6997245179063363,
|
| 1409 |
+
"grad_norm": 1.28125,
|
| 1410 |
+
"learning_rate": 4.921177276371069e-08,
|
| 1411 |
+
"loss": 1.6947,
|
| 1412 |
+
"step": 980
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 2.7134986225895315,
|
| 1416 |
+
"grad_norm": 1.2890625,
|
| 1417 |
+
"learning_rate": 4.483375932572597e-08,
|
| 1418 |
+
"loss": 1.6929,
|
| 1419 |
+
"step": 985
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 2.7272727272727275,
|
| 1423 |
+
"grad_norm": 1.2890625,
|
| 1424 |
+
"learning_rate": 4.0655204169127156e-08,
|
| 1425 |
+
"loss": 1.6944,
|
| 1426 |
+
"step": 990
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"epoch": 2.7410468319559227,
|
| 1430 |
+
"grad_norm": 1.2734375,
|
| 1431 |
+
"learning_rate": 3.667697986184526e-08,
|
| 1432 |
+
"loss": 1.6898,
|
| 1433 |
+
"step": 995
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 2.7548209366391183,
|
| 1437 |
+
"grad_norm": 1.2578125,
|
| 1438 |
+
"learning_rate": 3.2899917138625055e-08,
|
| 1439 |
+
"loss": 1.7061,
|
| 1440 |
+
"step": 1000
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 2.768595041322314,
|
| 1444 |
+
"grad_norm": 1.3125,
|
| 1445 |
+
"learning_rate": 2.9324804727551055e-08,
|
| 1446 |
+
"loss": 1.6974,
|
| 1447 |
+
"step": 1005
|
| 1448 |
+
},
|
| 1449 |
+
{
|
| 1450 |
+
"epoch": 2.7823691460055096,
|
| 1451 |
+
"grad_norm": 1.2734375,
|
| 1452 |
+
"learning_rate": 2.5952389185344925e-08,
|
| 1453 |
+
"loss": 1.6892,
|
| 1454 |
+
"step": 1010
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 2.796143250688705,
|
| 1458 |
+
"grad_norm": 1.2421875,
|
| 1459 |
+
"learning_rate": 2.2783374741469186e-08,
|
| 1460 |
+
"loss": 1.696,
|
| 1461 |
+
"step": 1015
|
| 1462 |
+
},
|
| 1463 |
+
{
|
| 1464 |
+
"epoch": 2.809917355371901,
|
| 1465 |
+
"grad_norm": 1.28125,
|
| 1466 |
+
"learning_rate": 1.9818423151069406e-08,
|
| 1467 |
+
"loss": 1.6879,
|
| 1468 |
+
"step": 1020
|
| 1469 |
+
},
|
| 1470 |
+
{
|
| 1471 |
+
"epoch": 2.8236914600550964,
|
| 1472 |
+
"grad_norm": 1.3125,
|
| 1473 |
+
"learning_rate": 1.705815355678619e-08,
|
| 1474 |
+
"loss": 1.6943,
|
| 1475 |
+
"step": 1025
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 2.837465564738292,
|
| 1479 |
+
"grad_norm": 1.3046875,
|
| 1480 |
+
"learning_rate": 1.4503142359465925e-08,
|
| 1481 |
+
"loss": 1.6919,
|
| 1482 |
+
"step": 1030
|
| 1483 |
+
},
|
| 1484 |
+
{
|
| 1485 |
+
"epoch": 2.8512396694214877,
|
| 1486 |
+
"grad_norm": 1.2421875,
|
| 1487 |
+
"learning_rate": 1.215392309779617e-08,
|
| 1488 |
+
"loss": 1.6907,
|
| 1489 |
+
"step": 1035
|
| 1490 |
+
},
|
| 1491 |
+
{
|
| 1492 |
+
"epoch": 2.8650137741046833,
|
| 1493 |
+
"grad_norm": 1.2734375,
|
| 1494 |
+
"learning_rate": 1.0010986336891458e-08,
|
| 1495 |
+
"loss": 1.704,
|
| 1496 |
+
"step": 1040
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 2.878787878787879,
|
| 1500 |
+
"grad_norm": 1.2578125,
|
| 1501 |
+
"learning_rate": 8.074779565854117e-09,
|
| 1502 |
+
"loss": 1.691,
|
| 1503 |
+
"step": 1045
|
| 1504 |
+
},
|
| 1505 |
+
{
|
| 1506 |
+
"epoch": 2.8925619834710745,
|
| 1507 |
+
"grad_norm": 1.3203125,
|
| 1508 |
+
"learning_rate": 6.34570710432869e-09,
|
| 1509 |
+
"loss": 1.6975,
|
| 1510 |
+
"step": 1050
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 2.90633608815427,
|
| 1514 |
+
"grad_norm": 1.2734375,
|
| 1515 |
+
"learning_rate": 4.824130018072026e-09,
|
| 1516 |
+
"loss": 1.6918,
|
| 1517 |
+
"step": 1055
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 2.9201101928374653,
|
| 1521 |
+
"grad_norm": 1.2890625,
|
| 1522 |
+
"learning_rate": 3.5103660435551465e-09,
|
| 1523 |
+
"loss": 1.6933,
|
| 1524 |
+
"step": 1060
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 2.9338842975206614,
|
| 1528 |
+
"grad_norm": 1.3046875,
|
| 1529 |
+
"learning_rate": 2.4046895216136563e-09,
|
| 1530 |
+
"loss": 1.6872,
|
| 1531 |
+
"step": 1065
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 2.9476584022038566,
|
| 1535 |
+
"grad_norm": 1.28125,
|
| 1536 |
+
"learning_rate": 1.5073313401594568e-09,
|
| 1537 |
+
"loss": 1.696,
|
| 1538 |
+
"step": 1070
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 2.9614325068870526,
|
| 1542 |
+
"grad_norm": 1.234375,
|
| 1543 |
+
"learning_rate": 8.184788859667557e-10,
|
| 1544 |
+
"loss": 1.6964,
|
| 1545 |
+
"step": 1075
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 2.975206611570248,
|
| 1549 |
+
"grad_norm": 1.3203125,
|
| 1550 |
+
"learning_rate": 3.3827600554170444e-10,
|
| 1551 |
+
"loss": 1.6941,
|
| 1552 |
+
"step": 1080
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 2.9889807162534434,
|
| 1556 |
+
"grad_norm": 1.3125,
|
| 1557 |
+
"learning_rate": 6.682297508464608e-11,
|
| 1558 |
+
"loss": 1.6993,
|
| 1559 |
+
"step": 1085
|
| 1560 |
}
|
| 1561 |
],
|
| 1562 |
"logging_steps": 5,
|
|
|
|
| 1571 |
"should_evaluate": false,
|
| 1572 |
"should_log": false,
|
| 1573 |
"should_save": true,
|
| 1574 |
+
"should_training_stop": true
|
| 1575 |
},
|
| 1576 |
"attributes": {}
|
| 1577 |
}
|
| 1578 |
},
|
| 1579 |
+
"total_flos": 6.427401199279931e+18,
|
| 1580 |
"train_batch_size": 8,
|
| 1581 |
"trial_name": null,
|
| 1582 |
"trial_params": null
|