Upload checkpoint-10000
Browse files- .gitattributes +18 -0
- optimizer_0/.metadata +3 -0
- optimizer_0/__0_0.distcp +3 -0
- optimizer_0/__1_0.distcp +3 -0
- optimizer_0/__2_0.distcp +3 -0
- optimizer_0/__3_0.distcp +3 -0
- optimizer_0/__4_0.distcp +3 -0
- optimizer_0/__5_0.distcp +3 -0
- optimizer_0/__6_0.distcp +3 -0
- optimizer_0/__7_0.distcp +3 -0
- pytorch_model_fsdp_0/.metadata +3 -0
- pytorch_model_fsdp_0/__0_0.distcp +3 -0
- pytorch_model_fsdp_0/__1_0.distcp +3 -0
- pytorch_model_fsdp_0/__2_0.distcp +3 -0
- pytorch_model_fsdp_0/__3_0.distcp +3 -0
- pytorch_model_fsdp_0/__4_0.distcp +3 -0
- pytorch_model_fsdp_0/__5_0.distcp +3 -0
- pytorch_model_fsdp_0/__6_0.distcp +3 -0
- pytorch_model_fsdp_0/__7_0.distcp +3 -0
- rng_state_0.pth +3 -0
- rng_state_1.pth +3 -0
- rng_state_2.pth +3 -0
- rng_state_3.pth +3 -0
- rng_state_4.pth +3 -0
- rng_state_5.pth +3 -0
- rng_state_6.pth +3 -0
- rng_state_7.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1434 -0
.gitattributes
CHANGED
|
@@ -51,3 +51,21 @@ checkpoint-30000/pytorch_model_fsdp_0/__4_0.distcp filter=lfs diff=lfs merge=lfs
|
|
| 51 |
checkpoint-30000/pytorch_model_fsdp_0/__5_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 52 |
checkpoint-30000/pytorch_model_fsdp_0/__6_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 53 |
checkpoint-30000/pytorch_model_fsdp_0/__7_0.distcp filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 51 |
checkpoint-30000/pytorch_model_fsdp_0/__5_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 52 |
checkpoint-30000/pytorch_model_fsdp_0/__6_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 53 |
checkpoint-30000/pytorch_model_fsdp_0/__7_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 54 |
+
optimizer_0/.metadata filter=lfs diff=lfs merge=lfs -text
|
| 55 |
+
optimizer_0/__0_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 56 |
+
optimizer_0/__1_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 57 |
+
optimizer_0/__2_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 58 |
+
optimizer_0/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 59 |
+
optimizer_0/__4_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 60 |
+
optimizer_0/__5_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 61 |
+
optimizer_0/__6_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 62 |
+
optimizer_0/__7_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 63 |
+
pytorch_model_fsdp_0/.metadata filter=lfs diff=lfs merge=lfs -text
|
| 64 |
+
pytorch_model_fsdp_0/__0_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 65 |
+
pytorch_model_fsdp_0/__1_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 66 |
+
pytorch_model_fsdp_0/__2_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 67 |
+
pytorch_model_fsdp_0/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 68 |
+
pytorch_model_fsdp_0/__4_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 69 |
+
pytorch_model_fsdp_0/__5_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 70 |
+
pytorch_model_fsdp_0/__6_0.distcp filter=lfs diff=lfs merge=lfs -text
|
| 71 |
+
pytorch_model_fsdp_0/__7_0.distcp filter=lfs diff=lfs merge=lfs -text
|
optimizer_0/.metadata
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0b6e4fef1449e764e63617312ff36b134b79f477317beb171b0740c69c31436f
|
| 3 |
+
size 1577571
|
optimizer_0/__0_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c0226fde4b561d4015ac6313a277611418c823493c39849a88ba84633d06268e
|
| 3 |
+
size 4042716233
|
optimizer_0/__1_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2d7117efae903d19712e435aa19fffaa0cbead5338e6d9f4fa5b4149b9ed4d76
|
| 3 |
+
size 4042809589
|
optimizer_0/__2_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2fcf70a3afc0eecc5a3c422cdf9a9253524c9150623838f66f330add51900465
|
| 3 |
+
size 4042809589
|
optimizer_0/__3_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c2ba060b65d30766479cd46064cb947192acc911f14dc1558b25b131d7714cde
|
| 3 |
+
size 4042809589
|
optimizer_0/__4_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4a0b3b750691b27820671ddc2ab53384d685d626d091863097fc52cab6505018
|
| 3 |
+
size 4042809589
|
optimizer_0/__5_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:86c726ccb1705178ea6c622e4e6dd6586559904e81f98ae1a0f384adf5e2da1a
|
| 3 |
+
size 4042817013
|
optimizer_0/__6_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e9d6ca7e9093e07dae5173cb196b143fe8300d0f9c2b19b472ecd16287d3b152
|
| 3 |
+
size 4042820552
|
optimizer_0/__7_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f9305f9819bde82082631043f69a533f745ce1d52538a7be0a70a05c0f9b5425
|
| 3 |
+
size 4042787848
|
pytorch_model_fsdp_0/.metadata
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d89018facf69483ec8a883be1c49b14d7509b34fab093f559511e3918103735a
|
| 3 |
+
size 1378806
|
pytorch_model_fsdp_0/__0_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ed809cb64b44279cb3633fb989bab302965db59f28c9b6070e05e81f31ea371c
|
| 3 |
+
size 2534754104
|
pytorch_model_fsdp_0/__1_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:48162aa26ea9cdd599615cda3a11128217371ce9b6ad6101f478114e6ccb6bff
|
| 3 |
+
size 2534754104
|
pytorch_model_fsdp_0/__2_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5d8c64fc91f736a3b33dcc19aff66190847210255eae743938de4049df2fc126
|
| 3 |
+
size 2534754104
|
pytorch_model_fsdp_0/__3_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5dc059c8e3b1e01ce76afc00f4db3b41b3cdc849838de5b35746fca8bddcb3f3
|
| 3 |
+
size 2534754104
|
pytorch_model_fsdp_0/__4_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bc80c4bfc63622c506486115b4c8bb14a5e289813fa6ef90a3a44b60dd345cc4
|
| 3 |
+
size 2534754104
|
pytorch_model_fsdp_0/__5_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cb6aa8e98b556a36ce9899d65e1698e91a04474222ea454c4f8066f4584bfb2d
|
| 3 |
+
size 2534754104
|
pytorch_model_fsdp_0/__6_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4c61fe1c477ab7bd3929b938aef60408f5e06d35ab6d62413ba69e11a1364f99
|
| 3 |
+
size 2534754104
|
pytorch_model_fsdp_0/__7_0.distcp
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d667d11364675a52a6728444ebc600e09e94734cc6cb6204bde60b104b8c13a8
|
| 3 |
+
size 2534713144
|
rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:826997d456e99c919bad937116b8aa018624d3cee7e87a68e844e70bd80aeb45
|
| 3 |
+
size 16389
|
rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:91ed4ea9de6889ae5075a8093b9c23505ab8c5554ee18ff06fea428794e1db0c
|
| 3 |
+
size 16389
|
rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a61500ed6f4ed1f30ad9f7e360264248052865b80b29cbdbe42cea3fcfdc3934
|
| 3 |
+
size 16389
|
rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:52136e6310f738523ce8b1da6f4ca37ccc5b3faa16f96d90873ade48d6d36b0a
|
| 3 |
+
size 16389
|
rng_state_4.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b2e4694a733324ad8470e895f12f022f928d19a23a5a2eacad0a568da191db37
|
| 3 |
+
size 16389
|
rng_state_5.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4ba38f72004469e5394586da18f61e173aa969dfa837593fc8e5b8247877b4f4
|
| 3 |
+
size 16389
|
rng_state_6.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6312012a936a01fbcbe06ee9ff87f68144ff51959bf530f08708c13ec6d49bc8
|
| 3 |
+
size 16389
|
rng_state_7.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51b2169702236d42a74f3bb15eff84b84fce6da260a8c333c66459901c75a30c
|
| 3 |
+
size 16389
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1a5a7b2dad441e5e2dccf21c3129e93bb2c435799b9325dae010bd3fda0c38e8
|
| 3 |
+
size 1465
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1434 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.331707964308223,
|
| 6 |
+
"eval_steps": 15000000,
|
| 7 |
+
"global_step": 10000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0016585398215411152,
|
| 14 |
+
"grad_norm": 5.2646002769470215,
|
| 15 |
+
"learning_rate": 1.9503781345362876e-06,
|
| 16 |
+
"loss": 3.606,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.0033170796430822304,
|
| 21 |
+
"grad_norm": 11.763270378112793,
|
| 22 |
+
"learning_rate": 3.940559904471274e-06,
|
| 23 |
+
"loss": 2.6238,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.004975619464623346,
|
| 28 |
+
"grad_norm": 6.007935047149658,
|
| 29 |
+
"learning_rate": 5.930741674406262e-06,
|
| 30 |
+
"loss": 2.5644,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.006634159286164461,
|
| 35 |
+
"grad_norm": 13.862035751342773,
|
| 36 |
+
"learning_rate": 7.920923444341248e-06,
|
| 37 |
+
"loss": 2.4601,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.008292699107705576,
|
| 42 |
+
"grad_norm": 6.1305341720581055,
|
| 43 |
+
"learning_rate": 9.911105214276236e-06,
|
| 44 |
+
"loss": 2.4099,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.009951238929246691,
|
| 49 |
+
"grad_norm": 9.804327964782715,
|
| 50 |
+
"learning_rate": 1.1901286984211225e-05,
|
| 51 |
+
"loss": 2.3409,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.011609778750787806,
|
| 56 |
+
"grad_norm": 6.416324138641357,
|
| 57 |
+
"learning_rate": 1.3891468754146211e-05,
|
| 58 |
+
"loss": 2.2951,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.013268318572328922,
|
| 63 |
+
"grad_norm": 9.413233757019043,
|
| 64 |
+
"learning_rate": 1.5881650524081196e-05,
|
| 65 |
+
"loss": 2.0864,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.014926858393870037,
|
| 70 |
+
"grad_norm": 4.721926212310791,
|
| 71 |
+
"learning_rate": 1.7871832294016186e-05,
|
| 72 |
+
"loss": 0.7346,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.016585398215411152,
|
| 77 |
+
"grad_norm": 3.284379720687866,
|
| 78 |
+
"learning_rate": 1.9862014063951173e-05,
|
| 79 |
+
"loss": 0.4516,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.01824393803695227,
|
| 84 |
+
"grad_norm": 2.671109199523926,
|
| 85 |
+
"learning_rate": 2.185219583388616e-05,
|
| 86 |
+
"loss": 0.3903,
|
| 87 |
+
"step": 550
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.019902477858493382,
|
| 91 |
+
"grad_norm": 2.644299268722534,
|
| 92 |
+
"learning_rate": 2.384237760382115e-05,
|
| 93 |
+
"loss": 0.3512,
|
| 94 |
+
"step": 600
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.0215610176800345,
|
| 98 |
+
"grad_norm": 2.1769394874572754,
|
| 99 |
+
"learning_rate": 2.5832559373756132e-05,
|
| 100 |
+
"loss": 0.3298,
|
| 101 |
+
"step": 650
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.023219557501575613,
|
| 105 |
+
"grad_norm": 2.0835165977478027,
|
| 106 |
+
"learning_rate": 2.7822741143691122e-05,
|
| 107 |
+
"loss": 0.3162,
|
| 108 |
+
"step": 700
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.02487809732311673,
|
| 112 |
+
"grad_norm": 2.069329023361206,
|
| 113 |
+
"learning_rate": 2.981292291362611e-05,
|
| 114 |
+
"loss": 0.3185,
|
| 115 |
+
"step": 750
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.026536637144657843,
|
| 119 |
+
"grad_norm": 1.8482165336608887,
|
| 120 |
+
"learning_rate": 3.1803104683561096e-05,
|
| 121 |
+
"loss": 0.3057,
|
| 122 |
+
"step": 800
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.02819517696619896,
|
| 126 |
+
"grad_norm": 1.8083229064941406,
|
| 127 |
+
"learning_rate": 3.379328645349608e-05,
|
| 128 |
+
"loss": 0.2857,
|
| 129 |
+
"step": 850
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.029853716787740073,
|
| 133 |
+
"grad_norm": 1.7274185419082642,
|
| 134 |
+
"learning_rate": 3.578346822343107e-05,
|
| 135 |
+
"loss": 0.3166,
|
| 136 |
+
"step": 900
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.03151225660928119,
|
| 140 |
+
"grad_norm": 1.6723179817199707,
|
| 141 |
+
"learning_rate": 3.777364999336606e-05,
|
| 142 |
+
"loss": 0.2981,
|
| 143 |
+
"step": 950
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.033170796430822304,
|
| 147 |
+
"grad_norm": 1.9198503494262695,
|
| 148 |
+
"learning_rate": 3.976383176330104e-05,
|
| 149 |
+
"loss": 0.2818,
|
| 150 |
+
"step": 1000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.03482933625236342,
|
| 154 |
+
"grad_norm": 1.680993914604187,
|
| 155 |
+
"learning_rate": 4.1754013533236035e-05,
|
| 156 |
+
"loss": 0.2909,
|
| 157 |
+
"step": 1050
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.03648787607390454,
|
| 161 |
+
"grad_norm": 1.6936819553375244,
|
| 162 |
+
"learning_rate": 4.374419530317102e-05,
|
| 163 |
+
"loss": 0.2871,
|
| 164 |
+
"step": 1100
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.03814641589544565,
|
| 168 |
+
"grad_norm": 1.460769534111023,
|
| 169 |
+
"learning_rate": 4.5734377073106e-05,
|
| 170 |
+
"loss": 0.2734,
|
| 171 |
+
"step": 1150
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.039804955716986765,
|
| 175 |
+
"grad_norm": 1.5563299655914307,
|
| 176 |
+
"learning_rate": 4.7724558843040995e-05,
|
| 177 |
+
"loss": 0.2883,
|
| 178 |
+
"step": 1200
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.04146349553852788,
|
| 182 |
+
"grad_norm": 1.4941402673721313,
|
| 183 |
+
"learning_rate": 4.971474061297598e-05,
|
| 184 |
+
"loss": 0.2858,
|
| 185 |
+
"step": 1250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.043122035360069,
|
| 189 |
+
"grad_norm": 1.2444454431533813,
|
| 190 |
+
"learning_rate": 5.170492238291097e-05,
|
| 191 |
+
"loss": 0.2626,
|
| 192 |
+
"step": 1300
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.04478057518161011,
|
| 196 |
+
"grad_norm": 1.2899850606918335,
|
| 197 |
+
"learning_rate": 5.3695104152845955e-05,
|
| 198 |
+
"loss": 0.29,
|
| 199 |
+
"step": 1350
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.046439115003151225,
|
| 203 |
+
"grad_norm": 1.6327451467514038,
|
| 204 |
+
"learning_rate": 5.568528592278094e-05,
|
| 205 |
+
"loss": 0.2897,
|
| 206 |
+
"step": 1400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.04809765482469234,
|
| 210 |
+
"grad_norm": 1.2772693634033203,
|
| 211 |
+
"learning_rate": 5.767546769271593e-05,
|
| 212 |
+
"loss": 0.2926,
|
| 213 |
+
"step": 1450
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.04975619464623346,
|
| 217 |
+
"grad_norm": 1.2572505474090576,
|
| 218 |
+
"learning_rate": 5.966564946265092e-05,
|
| 219 |
+
"loss": 0.2816,
|
| 220 |
+
"step": 1500
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.05141473446777457,
|
| 224 |
+
"grad_norm": 1.1974629163742065,
|
| 225 |
+
"learning_rate": 6.16558312325859e-05,
|
| 226 |
+
"loss": 0.2772,
|
| 227 |
+
"step": 1550
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.053073274289315686,
|
| 231 |
+
"grad_norm": 1.1929668188095093,
|
| 232 |
+
"learning_rate": 6.364601300252089e-05,
|
| 233 |
+
"loss": 0.289,
|
| 234 |
+
"step": 1600
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.0547318141108568,
|
| 238 |
+
"grad_norm": 1.2904795408248901,
|
| 239 |
+
"learning_rate": 6.563619477245587e-05,
|
| 240 |
+
"loss": 0.2792,
|
| 241 |
+
"step": 1650
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.05639035393239792,
|
| 245 |
+
"grad_norm": 1.169552206993103,
|
| 246 |
+
"learning_rate": 6.762637654239086e-05,
|
| 247 |
+
"loss": 0.2814,
|
| 248 |
+
"step": 1700
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.05804889375393903,
|
| 252 |
+
"grad_norm": 1.1359589099884033,
|
| 253 |
+
"learning_rate": 6.961655831232586e-05,
|
| 254 |
+
"loss": 0.2936,
|
| 255 |
+
"step": 1750
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.05970743357548015,
|
| 259 |
+
"grad_norm": 1.2144523859024048,
|
| 260 |
+
"learning_rate": 7.160674008226085e-05,
|
| 261 |
+
"loss": 0.272,
|
| 262 |
+
"step": 1800
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.06136597339702126,
|
| 266 |
+
"grad_norm": 1.1847233772277832,
|
| 267 |
+
"learning_rate": 7.359692185219582e-05,
|
| 268 |
+
"loss": 0.2635,
|
| 269 |
+
"step": 1850
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.06302451321856238,
|
| 273 |
+
"grad_norm": 1.231399655342102,
|
| 274 |
+
"learning_rate": 7.558710362213082e-05,
|
| 275 |
+
"loss": 0.2772,
|
| 276 |
+
"step": 1900
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.0646830530401035,
|
| 280 |
+
"grad_norm": 1.0221542119979858,
|
| 281 |
+
"learning_rate": 7.757728539206581e-05,
|
| 282 |
+
"loss": 0.2694,
|
| 283 |
+
"step": 1950
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.06634159286164461,
|
| 287 |
+
"grad_norm": 1.091098427772522,
|
| 288 |
+
"learning_rate": 7.956746716200078e-05,
|
| 289 |
+
"loss": 0.289,
|
| 290 |
+
"step": 2000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.06800013268318572,
|
| 294 |
+
"grad_norm": 1.1065690517425537,
|
| 295 |
+
"learning_rate": 8.155764893193577e-05,
|
| 296 |
+
"loss": 0.2974,
|
| 297 |
+
"step": 2050
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.06965867250472683,
|
| 301 |
+
"grad_norm": 0.9319403767585754,
|
| 302 |
+
"learning_rate": 8.354783070187077e-05,
|
| 303 |
+
"loss": 0.2819,
|
| 304 |
+
"step": 2100
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.07131721232626795,
|
| 308 |
+
"grad_norm": 1.1582205295562744,
|
| 309 |
+
"learning_rate": 8.553801247180575e-05,
|
| 310 |
+
"loss": 0.276,
|
| 311 |
+
"step": 2150
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.07297575214780908,
|
| 315 |
+
"grad_norm": 0.9972165822982788,
|
| 316 |
+
"learning_rate": 8.752819424174074e-05,
|
| 317 |
+
"loss": 0.2814,
|
| 318 |
+
"step": 2200
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.07463429196935019,
|
| 322 |
+
"grad_norm": 1.2905242443084717,
|
| 323 |
+
"learning_rate": 8.951837601167573e-05,
|
| 324 |
+
"loss": 0.2688,
|
| 325 |
+
"step": 2250
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.0762928317908913,
|
| 329 |
+
"grad_norm": 0.944217324256897,
|
| 330 |
+
"learning_rate": 9.150855778161071e-05,
|
| 331 |
+
"loss": 0.2786,
|
| 332 |
+
"step": 2300
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.07795137161243242,
|
| 336 |
+
"grad_norm": 0.9599469900131226,
|
| 337 |
+
"learning_rate": 9.349873955154569e-05,
|
| 338 |
+
"loss": 0.2677,
|
| 339 |
+
"step": 2350
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.07960991143397353,
|
| 343 |
+
"grad_norm": 1.0232480764389038,
|
| 344 |
+
"learning_rate": 9.54889213214807e-05,
|
| 345 |
+
"loss": 0.2751,
|
| 346 |
+
"step": 2400
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.08126845125551464,
|
| 350 |
+
"grad_norm": 0.8944919109344482,
|
| 351 |
+
"learning_rate": 9.747910309141567e-05,
|
| 352 |
+
"loss": 0.2883,
|
| 353 |
+
"step": 2450
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.08292699107705576,
|
| 357 |
+
"grad_norm": 0.8229078650474548,
|
| 358 |
+
"learning_rate": 9.946928486135066e-05,
|
| 359 |
+
"loss": 0.2783,
|
| 360 |
+
"step": 2500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.08458553089859687,
|
| 364 |
+
"grad_norm": 1.0346490144729614,
|
| 365 |
+
"learning_rate": 0.00010145946663128565,
|
| 366 |
+
"loss": 0.2774,
|
| 367 |
+
"step": 2550
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.086244070720138,
|
| 371 |
+
"grad_norm": 1.1092647314071655,
|
| 372 |
+
"learning_rate": 0.00010344964840122063,
|
| 373 |
+
"loss": 0.254,
|
| 374 |
+
"step": 2600
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.08790261054167911,
|
| 378 |
+
"grad_norm": 1.0459065437316895,
|
| 379 |
+
"learning_rate": 0.00010543983017115562,
|
| 380 |
+
"loss": 0.2907,
|
| 381 |
+
"step": 2650
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.08956115036322022,
|
| 385 |
+
"grad_norm": 0.8515977263450623,
|
| 386 |
+
"learning_rate": 0.00010743001194109062,
|
| 387 |
+
"loss": 0.2767,
|
| 388 |
+
"step": 2700
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.09121969018476134,
|
| 392 |
+
"grad_norm": 0.8556401133537292,
|
| 393 |
+
"learning_rate": 0.0001094201937110256,
|
| 394 |
+
"loss": 0.2657,
|
| 395 |
+
"step": 2750
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.09287823000630245,
|
| 399 |
+
"grad_norm": 0.9892629384994507,
|
| 400 |
+
"learning_rate": 0.00011141037548096058,
|
| 401 |
+
"loss": 0.2635,
|
| 402 |
+
"step": 2800
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.09453676982784356,
|
| 406 |
+
"grad_norm": 0.8110876679420471,
|
| 407 |
+
"learning_rate": 0.00011340055725089557,
|
| 408 |
+
"loss": 0.2783,
|
| 409 |
+
"step": 2850
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.09619530964938468,
|
| 413 |
+
"grad_norm": 0.9971825480461121,
|
| 414 |
+
"learning_rate": 0.00011539073902083055,
|
| 415 |
+
"loss": 0.2723,
|
| 416 |
+
"step": 2900
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.09785384947092579,
|
| 420 |
+
"grad_norm": 0.8159605860710144,
|
| 421 |
+
"learning_rate": 0.00011738092079076555,
|
| 422 |
+
"loss": 0.2904,
|
| 423 |
+
"step": 2950
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.09951238929246692,
|
| 427 |
+
"grad_norm": 0.8371638655662537,
|
| 428 |
+
"learning_rate": 0.00011937110256070054,
|
| 429 |
+
"loss": 0.2725,
|
| 430 |
+
"step": 3000
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.10117092911400803,
|
| 434 |
+
"grad_norm": 0.9144733548164368,
|
| 435 |
+
"learning_rate": 0.00012136128433063553,
|
| 436 |
+
"loss": 0.2733,
|
| 437 |
+
"step": 3050
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.10282946893554915,
|
| 441 |
+
"grad_norm": 0.9184954762458801,
|
| 442 |
+
"learning_rate": 0.0001233514661005705,
|
| 443 |
+
"loss": 0.2778,
|
| 444 |
+
"step": 3100
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.10448800875709026,
|
| 448 |
+
"grad_norm": 0.760306179523468,
|
| 449 |
+
"learning_rate": 0.00012534164787050549,
|
| 450 |
+
"loss": 0.2714,
|
| 451 |
+
"step": 3150
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.10614654857863137,
|
| 455 |
+
"grad_norm": 0.7679548263549805,
|
| 456 |
+
"learning_rate": 0.00012733182964044047,
|
| 457 |
+
"loss": 0.2607,
|
| 458 |
+
"step": 3200
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.10780508840017249,
|
| 462 |
+
"grad_norm": 0.899803876876831,
|
| 463 |
+
"learning_rate": 0.00012932201141037549,
|
| 464 |
+
"loss": 0.2572,
|
| 465 |
+
"step": 3250
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.1094636282217136,
|
| 469 |
+
"grad_norm": 0.7616294026374817,
|
| 470 |
+
"learning_rate": 0.00013131219318031047,
|
| 471 |
+
"loss": 0.2733,
|
| 472 |
+
"step": 3300
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.11112216804325471,
|
| 476 |
+
"grad_norm": 0.8336801528930664,
|
| 477 |
+
"learning_rate": 0.00013330237495024543,
|
| 478 |
+
"loss": 0.274,
|
| 479 |
+
"step": 3350
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.11278070786479584,
|
| 483 |
+
"grad_norm": 0.7640217542648315,
|
| 484 |
+
"learning_rate": 0.00013529255672018042,
|
| 485 |
+
"loss": 0.2784,
|
| 486 |
+
"step": 3400
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.11443924768633695,
|
| 490 |
+
"grad_norm": 0.7056282758712769,
|
| 491 |
+
"learning_rate": 0.0001372827384901154,
|
| 492 |
+
"loss": 0.2639,
|
| 493 |
+
"step": 3450
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.11609778750787807,
|
| 497 |
+
"grad_norm": 0.8658528327941895,
|
| 498 |
+
"learning_rate": 0.00013927292026005042,
|
| 499 |
+
"loss": 0.2681,
|
| 500 |
+
"step": 3500
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.11775632732941918,
|
| 504 |
+
"grad_norm": 0.802803099155426,
|
| 505 |
+
"learning_rate": 0.0001412631020299854,
|
| 506 |
+
"loss": 0.2963,
|
| 507 |
+
"step": 3550
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.1194148671509603,
|
| 511 |
+
"grad_norm": 0.8917680978775024,
|
| 512 |
+
"learning_rate": 0.0001432532837999204,
|
| 513 |
+
"loss": 0.2826,
|
| 514 |
+
"step": 3600
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.12107340697250141,
|
| 518 |
+
"grad_norm": 0.7213064432144165,
|
| 519 |
+
"learning_rate": 0.00014524346556985538,
|
| 520 |
+
"loss": 0.2758,
|
| 521 |
+
"step": 3650
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.12273194679404252,
|
| 525 |
+
"grad_norm": 0.6630449891090393,
|
| 526 |
+
"learning_rate": 0.00014723364733979034,
|
| 527 |
+
"loss": 0.2669,
|
| 528 |
+
"step": 3700
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.12439048661558363,
|
| 532 |
+
"grad_norm": 0.9187581539154053,
|
| 533 |
+
"learning_rate": 0.00014922382910972532,
|
| 534 |
+
"loss": 0.2664,
|
| 535 |
+
"step": 3750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.12604902643712476,
|
| 539 |
+
"grad_norm": 0.7035880088806152,
|
| 540 |
+
"learning_rate": 0.0001512140108796603,
|
| 541 |
+
"loss": 0.2813,
|
| 542 |
+
"step": 3800
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.12770756625866586,
|
| 546 |
+
"grad_norm": 0.9671415686607361,
|
| 547 |
+
"learning_rate": 0.0001532041926495953,
|
| 548 |
+
"loss": 0.2892,
|
| 549 |
+
"step": 3850
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.129366106080207,
|
| 553 |
+
"grad_norm": 0.809038519859314,
|
| 554 |
+
"learning_rate": 0.00015519437441953028,
|
| 555 |
+
"loss": 0.2662,
|
| 556 |
+
"step": 3900
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.1310246459017481,
|
| 560 |
+
"grad_norm": 0.8309035301208496,
|
| 561 |
+
"learning_rate": 0.0001571845561894653,
|
| 562 |
+
"loss": 0.2567,
|
| 563 |
+
"step": 3950
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.13268318572328922,
|
| 567 |
+
"grad_norm": 0.7343366146087646,
|
| 568 |
+
"learning_rate": 0.00015917473795940028,
|
| 569 |
+
"loss": 0.2789,
|
| 570 |
+
"step": 4000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.13434172554483034,
|
| 574 |
+
"grad_norm": 0.6523250341415405,
|
| 575 |
+
"learning_rate": 0.00016116491972933527,
|
| 576 |
+
"loss": 0.2568,
|
| 577 |
+
"step": 4050
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.13600026536637144,
|
| 581 |
+
"grad_norm": 0.6583340167999268,
|
| 582 |
+
"learning_rate": 0.00016315510149927026,
|
| 583 |
+
"loss": 0.2694,
|
| 584 |
+
"step": 4100
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.13765880518791257,
|
| 588 |
+
"grad_norm": 0.674410879611969,
|
| 589 |
+
"learning_rate": 0.00016514528326920524,
|
| 590 |
+
"loss": 0.2839,
|
| 591 |
+
"step": 4150
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.13931734500945367,
|
| 595 |
+
"grad_norm": 0.9803613424301147,
|
| 596 |
+
"learning_rate": 0.00016713546503914023,
|
| 597 |
+
"loss": 0.2789,
|
| 598 |
+
"step": 4200
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.1409758848309948,
|
| 602 |
+
"grad_norm": 0.6113103032112122,
|
| 603 |
+
"learning_rate": 0.00016912564680907522,
|
| 604 |
+
"loss": 0.263,
|
| 605 |
+
"step": 4250
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.1426344246525359,
|
| 609 |
+
"grad_norm": 0.6843391060829163,
|
| 610 |
+
"learning_rate": 0.0001711158285790102,
|
| 611 |
+
"loss": 0.2673,
|
| 612 |
+
"step": 4300
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.14429296447407702,
|
| 616 |
+
"grad_norm": 0.7611909508705139,
|
| 617 |
+
"learning_rate": 0.0001731060103489452,
|
| 618 |
+
"loss": 0.2628,
|
| 619 |
+
"step": 4350
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.14595150429561815,
|
| 623 |
+
"grad_norm": 0.7338739037513733,
|
| 624 |
+
"learning_rate": 0.00017509619211888015,
|
| 625 |
+
"loss": 0.2728,
|
| 626 |
+
"step": 4400
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.14761004411715925,
|
| 630 |
+
"grad_norm": 0.6578771471977234,
|
| 631 |
+
"learning_rate": 0.00017708637388881514,
|
| 632 |
+
"loss": 0.2503,
|
| 633 |
+
"step": 4450
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.14926858393870038,
|
| 637 |
+
"grad_norm": 0.6933615803718567,
|
| 638 |
+
"learning_rate": 0.00017907655565875018,
|
| 639 |
+
"loss": 0.2655,
|
| 640 |
+
"step": 4500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.15092712376024148,
|
| 644 |
+
"grad_norm": 0.574677586555481,
|
| 645 |
+
"learning_rate": 0.00018106673742868514,
|
| 646 |
+
"loss": 0.2545,
|
| 647 |
+
"step": 4550
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.1525856635817826,
|
| 651 |
+
"grad_norm": 0.5872885584831238,
|
| 652 |
+
"learning_rate": 0.00018305691919862012,
|
| 653 |
+
"loss": 0.2728,
|
| 654 |
+
"step": 4600
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.1542442034033237,
|
| 658 |
+
"grad_norm": 0.7030934691429138,
|
| 659 |
+
"learning_rate": 0.0001850471009685551,
|
| 660 |
+
"loss": 0.2691,
|
| 661 |
+
"step": 4650
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.15590274322486483,
|
| 665 |
+
"grad_norm": 0.6045774817466736,
|
| 666 |
+
"learning_rate": 0.0001870372827384901,
|
| 667 |
+
"loss": 0.2676,
|
| 668 |
+
"step": 4700
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.15756128304640593,
|
| 672 |
+
"grad_norm": 0.6607248783111572,
|
| 673 |
+
"learning_rate": 0.00018902746450842508,
|
| 674 |
+
"loss": 0.2706,
|
| 675 |
+
"step": 4750
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.15921982286794706,
|
| 679 |
+
"grad_norm": 0.6813399195671082,
|
| 680 |
+
"learning_rate": 0.00019101764627836007,
|
| 681 |
+
"loss": 0.2602,
|
| 682 |
+
"step": 4800
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.16087836268948819,
|
| 686 |
+
"grad_norm": 0.6294276714324951,
|
| 687 |
+
"learning_rate": 0.00019300782804829506,
|
| 688 |
+
"loss": 0.2861,
|
| 689 |
+
"step": 4850
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.16253690251102929,
|
| 693 |
+
"grad_norm": 0.7689523696899414,
|
| 694 |
+
"learning_rate": 0.00019499800981823004,
|
| 695 |
+
"loss": 0.2919,
|
| 696 |
+
"step": 4900
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.1641954423325704,
|
| 700 |
+
"grad_norm": 0.6293140053749084,
|
| 701 |
+
"learning_rate": 0.00019698819158816503,
|
| 702 |
+
"loss": 0.2625,
|
| 703 |
+
"step": 4950
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.1658539821541115,
|
| 707 |
+
"grad_norm": 0.6510013937950134,
|
| 708 |
+
"learning_rate": 0.00019897837335810002,
|
| 709 |
+
"loss": 0.2645,
|
| 710 |
+
"step": 5000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.16751252197565264,
|
| 714 |
+
"grad_norm": 0.655305027961731,
|
| 715 |
+
"learning_rate": 0.00020096855512803503,
|
| 716 |
+
"loss": 0.2899,
|
| 717 |
+
"step": 5050
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.16917106179719374,
|
| 721 |
+
"grad_norm": 0.6357027888298035,
|
| 722 |
+
"learning_rate": 0.00020295873689797002,
|
| 723 |
+
"loss": 0.2841,
|
| 724 |
+
"step": 5100
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.17082960161873487,
|
| 728 |
+
"grad_norm": 0.6427227258682251,
|
| 729 |
+
"learning_rate": 0.000204948918667905,
|
| 730 |
+
"loss": 0.2904,
|
| 731 |
+
"step": 5150
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.172488141440276,
|
| 735 |
+
"grad_norm": 0.5456770658493042,
|
| 736 |
+
"learning_rate": 0.00020693910043784,
|
| 737 |
+
"loss": 0.2606,
|
| 738 |
+
"step": 5200
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.1741466812618171,
|
| 742 |
+
"grad_norm": 0.6049277186393738,
|
| 743 |
+
"learning_rate": 0.00020892928220777495,
|
| 744 |
+
"loss": 0.2789,
|
| 745 |
+
"step": 5250
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.17580522108335822,
|
| 749 |
+
"grad_norm": 0.604241669178009,
|
| 750 |
+
"learning_rate": 0.00021091946397770994,
|
| 751 |
+
"loss": 0.2855,
|
| 752 |
+
"step": 5300
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.17746376090489932,
|
| 756 |
+
"grad_norm": 0.5764302015304565,
|
| 757 |
+
"learning_rate": 0.00021290964574764492,
|
| 758 |
+
"loss": 0.2724,
|
| 759 |
+
"step": 5350
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.17912230072644045,
|
| 763 |
+
"grad_norm": 0.5926195979118347,
|
| 764 |
+
"learning_rate": 0.0002148998275175799,
|
| 765 |
+
"loss": 0.2693,
|
| 766 |
+
"step": 5400
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.18078084054798155,
|
| 770 |
+
"grad_norm": 0.550534188747406,
|
| 771 |
+
"learning_rate": 0.0002168900092875149,
|
| 772 |
+
"loss": 0.2739,
|
| 773 |
+
"step": 5450
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.18243938036952267,
|
| 777 |
+
"grad_norm": 0.9231746792793274,
|
| 778 |
+
"learning_rate": 0.00021888019105744988,
|
| 779 |
+
"loss": 0.2713,
|
| 780 |
+
"step": 5500
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.18409792019106377,
|
| 784 |
+
"grad_norm": 0.6074857115745544,
|
| 785 |
+
"learning_rate": 0.00022087037282738487,
|
| 786 |
+
"loss": 0.2628,
|
| 787 |
+
"step": 5550
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.1857564600126049,
|
| 791 |
+
"grad_norm": 0.5123354196548462,
|
| 792 |
+
"learning_rate": 0.00022286055459731988,
|
| 793 |
+
"loss": 0.2752,
|
| 794 |
+
"step": 5600
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.18741499983414603,
|
| 798 |
+
"grad_norm": 0.5391439199447632,
|
| 799 |
+
"learning_rate": 0.00022485073636725487,
|
| 800 |
+
"loss": 0.2772,
|
| 801 |
+
"step": 5650
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.18907353965568713,
|
| 805 |
+
"grad_norm": 0.5493964552879333,
|
| 806 |
+
"learning_rate": 0.00022684091813718986,
|
| 807 |
+
"loss": 0.2656,
|
| 808 |
+
"step": 5700
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.19073207947722826,
|
| 812 |
+
"grad_norm": 0.531689465045929,
|
| 813 |
+
"learning_rate": 0.00022883109990712484,
|
| 814 |
+
"loss": 0.2652,
|
| 815 |
+
"step": 5750
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.19239061929876936,
|
| 819 |
+
"grad_norm": 0.6406611204147339,
|
| 820 |
+
"learning_rate": 0.00023082128167705983,
|
| 821 |
+
"loss": 0.2807,
|
| 822 |
+
"step": 5800
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.19404915912031048,
|
| 826 |
+
"grad_norm": 0.5147985219955444,
|
| 827 |
+
"learning_rate": 0.00023281146344699481,
|
| 828 |
+
"loss": 0.2859,
|
| 829 |
+
"step": 5850
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.19570769894185158,
|
| 833 |
+
"grad_norm": 0.5153710246086121,
|
| 834 |
+
"learning_rate": 0.0002348016452169298,
|
| 835 |
+
"loss": 0.2529,
|
| 836 |
+
"step": 5900
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.1973662387633927,
|
| 840 |
+
"grad_norm": 0.5277467966079712,
|
| 841 |
+
"learning_rate": 0.00023679182698686476,
|
| 842 |
+
"loss": 0.2731,
|
| 843 |
+
"step": 5950
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.19902477858493384,
|
| 847 |
+
"grad_norm": 0.480101615190506,
|
| 848 |
+
"learning_rate": 0.00023878200875679975,
|
| 849 |
+
"loss": 0.2784,
|
| 850 |
+
"step": 6000
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.20068331840647494,
|
| 854 |
+
"grad_norm": 0.5489948987960815,
|
| 855 |
+
"learning_rate": 0.00024077219052673473,
|
| 856 |
+
"loss": 0.2968,
|
| 857 |
+
"step": 6050
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.20234185822801606,
|
| 861 |
+
"grad_norm": 0.6148312091827393,
|
| 862 |
+
"learning_rate": 0.00024276237229666972,
|
| 863 |
+
"loss": 0.2731,
|
| 864 |
+
"step": 6100
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.20400039804955716,
|
| 868 |
+
"grad_norm": 0.5747624039649963,
|
| 869 |
+
"learning_rate": 0.00024475255406660476,
|
| 870 |
+
"loss": 0.2701,
|
| 871 |
+
"step": 6150
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.2056589378710983,
|
| 875 |
+
"grad_norm": 0.6135842800140381,
|
| 876 |
+
"learning_rate": 0.00024674273583653975,
|
| 877 |
+
"loss": 0.2691,
|
| 878 |
+
"step": 6200
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.2073174776926394,
|
| 882 |
+
"grad_norm": 0.5454925298690796,
|
| 883 |
+
"learning_rate": 0.00024873291760647473,
|
| 884 |
+
"loss": 0.2737,
|
| 885 |
+
"step": 6250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.20897601751418052,
|
| 889 |
+
"grad_norm": 0.6004371047019958,
|
| 890 |
+
"learning_rate": 0.0002507230993764097,
|
| 891 |
+
"loss": 0.288,
|
| 892 |
+
"step": 6300
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.21063455733572162,
|
| 896 |
+
"grad_norm": 0.5263451933860779,
|
| 897 |
+
"learning_rate": 0.0002527132811463447,
|
| 898 |
+
"loss": 0.2752,
|
| 899 |
+
"step": 6350
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.21229309715726274,
|
| 903 |
+
"grad_norm": 0.5507785677909851,
|
| 904 |
+
"learning_rate": 0.00025470346291627964,
|
| 905 |
+
"loss": 0.2749,
|
| 906 |
+
"step": 6400
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.21395163697880387,
|
| 910 |
+
"grad_norm": 0.5042358636856079,
|
| 911 |
+
"learning_rate": 0.0002566936446862146,
|
| 912 |
+
"loss": 0.2773,
|
| 913 |
+
"step": 6450
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.21561017680034497,
|
| 917 |
+
"grad_norm": 0.4615878164768219,
|
| 918 |
+
"learning_rate": 0.0002586838264561496,
|
| 919 |
+
"loss": 0.2782,
|
| 920 |
+
"step": 6500
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.2172687166218861,
|
| 924 |
+
"grad_norm": 0.5063010454177856,
|
| 925 |
+
"learning_rate": 0.0002606740082260846,
|
| 926 |
+
"loss": 0.2805,
|
| 927 |
+
"step": 6550
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.2189272564434272,
|
| 931 |
+
"grad_norm": 0.5367996096611023,
|
| 932 |
+
"learning_rate": 0.0002626641899960196,
|
| 933 |
+
"loss": 0.2689,
|
| 934 |
+
"step": 6600
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.22058579626496833,
|
| 938 |
+
"grad_norm": 0.4879344701766968,
|
| 939 |
+
"learning_rate": 0.0002646543717659546,
|
| 940 |
+
"loss": 0.2745,
|
| 941 |
+
"step": 6650
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.22224433608650943,
|
| 945 |
+
"grad_norm": 0.504068911075592,
|
| 946 |
+
"learning_rate": 0.00026664455353588956,
|
| 947 |
+
"loss": 0.2685,
|
| 948 |
+
"step": 6700
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.22390287590805055,
|
| 952 |
+
"grad_norm": 0.4813278913497925,
|
| 953 |
+
"learning_rate": 0.0002686347353058246,
|
| 954 |
+
"loss": 0.261,
|
| 955 |
+
"step": 6750
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.22556141572959168,
|
| 959 |
+
"grad_norm": 0.46839526295661926,
|
| 960 |
+
"learning_rate": 0.0002706249170757596,
|
| 961 |
+
"loss": 0.2702,
|
| 962 |
+
"step": 6800
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.22721995555113278,
|
| 966 |
+
"grad_norm": 0.5233504176139832,
|
| 967 |
+
"learning_rate": 0.0002726150988456946,
|
| 968 |
+
"loss": 0.2736,
|
| 969 |
+
"step": 6850
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.2288784953726739,
|
| 973 |
+
"grad_norm": 0.5277813673019409,
|
| 974 |
+
"learning_rate": 0.00027460528061562956,
|
| 975 |
+
"loss": 0.2772,
|
| 976 |
+
"step": 6900
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.230537035194215,
|
| 980 |
+
"grad_norm": 0.6029180884361267,
|
| 981 |
+
"learning_rate": 0.00027659546238556455,
|
| 982 |
+
"loss": 0.2768,
|
| 983 |
+
"step": 6950
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.23219557501575613,
|
| 987 |
+
"grad_norm": 0.575391948223114,
|
| 988 |
+
"learning_rate": 0.00027858564415549953,
|
| 989 |
+
"loss": 0.265,
|
| 990 |
+
"step": 7000
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.23385411483729723,
|
| 994 |
+
"grad_norm": 0.4373869299888611,
|
| 995 |
+
"learning_rate": 0.0002805758259254345,
|
| 996 |
+
"loss": 0.2657,
|
| 997 |
+
"step": 7050
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.23551265465883836,
|
| 1001 |
+
"grad_norm": 0.8708984851837158,
|
| 1002 |
+
"learning_rate": 0.00028256600769536945,
|
| 1003 |
+
"loss": 0.2754,
|
| 1004 |
+
"step": 7100
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.23717119448037946,
|
| 1008 |
+
"grad_norm": 0.5389964580535889,
|
| 1009 |
+
"learning_rate": 0.00028455618946530444,
|
| 1010 |
+
"loss": 0.2654,
|
| 1011 |
+
"step": 7150
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.2388297343019206,
|
| 1015 |
+
"grad_norm": 0.5265533328056335,
|
| 1016 |
+
"learning_rate": 0.0002865463712352394,
|
| 1017 |
+
"loss": 0.2943,
|
| 1018 |
+
"step": 7200
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.24048827412346171,
|
| 1022 |
+
"grad_norm": 0.43175622820854187,
|
| 1023 |
+
"learning_rate": 0.0002885365530051744,
|
| 1024 |
+
"loss": 0.2597,
|
| 1025 |
+
"step": 7250
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.24214681394500281,
|
| 1029 |
+
"grad_norm": 0.42523327469825745,
|
| 1030 |
+
"learning_rate": 0.00029052673477510945,
|
| 1031 |
+
"loss": 0.281,
|
| 1032 |
+
"step": 7300
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.24380535376654394,
|
| 1036 |
+
"grad_norm": 0.4567892551422119,
|
| 1037 |
+
"learning_rate": 0.00029251691654504444,
|
| 1038 |
+
"loss": 0.2776,
|
| 1039 |
+
"step": 7350
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.24546389358808504,
|
| 1043 |
+
"grad_norm": 0.4635256826877594,
|
| 1044 |
+
"learning_rate": 0.0002945070983149794,
|
| 1045 |
+
"loss": 0.2608,
|
| 1046 |
+
"step": 7400
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.24712243340962617,
|
| 1050 |
+
"grad_norm": 0.43223607540130615,
|
| 1051 |
+
"learning_rate": 0.0002964972800849144,
|
| 1052 |
+
"loss": 0.2577,
|
| 1053 |
+
"step": 7450
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.24878097323116727,
|
| 1057 |
+
"grad_norm": 0.4766380190849304,
|
| 1058 |
+
"learning_rate": 0.0002984874618548494,
|
| 1059 |
+
"loss": 0.2717,
|
| 1060 |
+
"step": 7500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.2504395130527084,
|
| 1064 |
+
"grad_norm": 0.47827085852622986,
|
| 1065 |
+
"learning_rate": 0.0002999999948018406,
|
| 1066 |
+
"loss": 0.2748,
|
| 1067 |
+
"step": 7550
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.2520980528742495,
|
| 1071 |
+
"grad_norm": 0.5068143606185913,
|
| 1072 |
+
"learning_rate": 0.000299999861238043,
|
| 1073 |
+
"loss": 0.2786,
|
| 1074 |
+
"step": 7600
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.25375659269579065,
|
| 1078 |
+
"grad_norm": 0.48091641068458557,
|
| 1079 |
+
"learning_rate": 0.0002999995471827833,
|
| 1080 |
+
"loss": 0.2623,
|
| 1081 |
+
"step": 7650
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.2554151325173317,
|
| 1085 |
+
"grad_norm": 0.5682414770126343,
|
| 1086 |
+
"learning_rate": 0.00029999905263643944,
|
| 1087 |
+
"loss": 0.2784,
|
| 1088 |
+
"step": 7700
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.25707367233887285,
|
| 1092 |
+
"grad_norm": 0.42487502098083496,
|
| 1093 |
+
"learning_rate": 0.00029999837759960636,
|
| 1094 |
+
"loss": 0.2632,
|
| 1095 |
+
"step": 7750
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.258732212160414,
|
| 1099 |
+
"grad_norm": 0.3851436674594879,
|
| 1100 |
+
"learning_rate": 0.00029999752207309643,
|
| 1101 |
+
"loss": 0.262,
|
| 1102 |
+
"step": 7800
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.2603907519819551,
|
| 1106 |
+
"grad_norm": 0.4501873552799225,
|
| 1107 |
+
"learning_rate": 0.000299996486057939,
|
| 1108 |
+
"loss": 0.2579,
|
| 1109 |
+
"step": 7850
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.2620492918034962,
|
| 1113 |
+
"grad_norm": 0.42192208766937256,
|
| 1114 |
+
"learning_rate": 0.00029999526955538073,
|
| 1115 |
+
"loss": 0.2651,
|
| 1116 |
+
"step": 7900
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.2637078316250373,
|
| 1120 |
+
"grad_norm": 0.3992311656475067,
|
| 1121 |
+
"learning_rate": 0.0002999938725668854,
|
| 1122 |
+
"loss": 0.2687,
|
| 1123 |
+
"step": 7950
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.26536637144657843,
|
| 1127 |
+
"grad_norm": 0.4475759267807007,
|
| 1128 |
+
"learning_rate": 0.000299992295094134,
|
| 1129 |
+
"loss": 0.262,
|
| 1130 |
+
"step": 8000
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.26702491126811956,
|
| 1134 |
+
"grad_norm": 1.0701886415481567,
|
| 1135 |
+
"learning_rate": 0.00029999053713902464,
|
| 1136 |
+
"loss": 0.2702,
|
| 1137 |
+
"step": 8050
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.2686834510896607,
|
| 1141 |
+
"grad_norm": 0.7080543637275696,
|
| 1142 |
+
"learning_rate": 0.00029998859870367263,
|
| 1143 |
+
"loss": 0.2752,
|
| 1144 |
+
"step": 8100
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.27034199091120176,
|
| 1148 |
+
"grad_norm": 0.47246304154396057,
|
| 1149 |
+
"learning_rate": 0.0002999864797904105,
|
| 1150 |
+
"loss": 0.2708,
|
| 1151 |
+
"step": 8150
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.2720005307327429,
|
| 1155 |
+
"grad_norm": 0.3701266348361969,
|
| 1156 |
+
"learning_rate": 0.0002999841804017878,
|
| 1157 |
+
"loss": 0.2607,
|
| 1158 |
+
"step": 8200
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.273659070554284,
|
| 1162 |
+
"grad_norm": 0.41229259967803955,
|
| 1163 |
+
"learning_rate": 0.00029998170054057135,
|
| 1164 |
+
"loss": 0.2618,
|
| 1165 |
+
"step": 8250
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.27531761037582514,
|
| 1169 |
+
"grad_norm": 0.41077643632888794,
|
| 1170 |
+
"learning_rate": 0.00029997904020974517,
|
| 1171 |
+
"loss": 0.2554,
|
| 1172 |
+
"step": 8300
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.27697615019736627,
|
| 1176 |
+
"grad_norm": 0.36561328172683716,
|
| 1177 |
+
"learning_rate": 0.0002999761994125103,
|
| 1178 |
+
"loss": 0.2764,
|
| 1179 |
+
"step": 8350
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.27863469001890734,
|
| 1183 |
+
"grad_norm": 0.4316375255584717,
|
| 1184 |
+
"learning_rate": 0.00029997317815228506,
|
| 1185 |
+
"loss": 0.2678,
|
| 1186 |
+
"step": 8400
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.28029322984044847,
|
| 1190 |
+
"grad_norm": 0.4210038483142853,
|
| 1191 |
+
"learning_rate": 0.0002999699764327049,
|
| 1192 |
+
"loss": 0.2462,
|
| 1193 |
+
"step": 8450
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.2819517696619896,
|
| 1197 |
+
"grad_norm": 0.4256073236465454,
|
| 1198 |
+
"learning_rate": 0.0002999665942576223,
|
| 1199 |
+
"loss": 0.271,
|
| 1200 |
+
"step": 8500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.2836103094835307,
|
| 1204 |
+
"grad_norm": 0.40567606687545776,
|
| 1205 |
+
"learning_rate": 0.000299963031631107,
|
| 1206 |
+
"loss": 0.2574,
|
| 1207 |
+
"step": 8550
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.2852688493050718,
|
| 1211 |
+
"grad_norm": 0.3660525381565094,
|
| 1212 |
+
"learning_rate": 0.00029995928855744577,
|
| 1213 |
+
"loss": 0.2697,
|
| 1214 |
+
"step": 8600
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.2869273891266129,
|
| 1218 |
+
"grad_norm": 0.44151097536087036,
|
| 1219 |
+
"learning_rate": 0.0002999553650411427,
|
| 1220 |
+
"loss": 0.2508,
|
| 1221 |
+
"step": 8650
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.28858592894815405,
|
| 1225 |
+
"grad_norm": 0.38013362884521484,
|
| 1226 |
+
"learning_rate": 0.00029995126108691865,
|
| 1227 |
+
"loss": 0.237,
|
| 1228 |
+
"step": 8700
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.2902444687696952,
|
| 1232 |
+
"grad_norm": 0.44534966349601746,
|
| 1233 |
+
"learning_rate": 0.00029994697669971204,
|
| 1234 |
+
"loss": 0.2575,
|
| 1235 |
+
"step": 8750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.2919030085912363,
|
| 1239 |
+
"grad_norm": 0.40814974904060364,
|
| 1240 |
+
"learning_rate": 0.00029994251188467813,
|
| 1241 |
+
"loss": 0.2494,
|
| 1242 |
+
"step": 8800
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.2935615484127774,
|
| 1246 |
+
"grad_norm": 0.3318258821964264,
|
| 1247 |
+
"learning_rate": 0.0002999378666471892,
|
| 1248 |
+
"loss": 0.2381,
|
| 1249 |
+
"step": 8850
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.2952200882343185,
|
| 1253 |
+
"grad_norm": 0.43739181756973267,
|
| 1254 |
+
"learning_rate": 0.00029993304099283483,
|
| 1255 |
+
"loss": 0.2388,
|
| 1256 |
+
"step": 8900
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.29687862805585963,
|
| 1260 |
+
"grad_norm": 0.3652569055557251,
|
| 1261 |
+
"learning_rate": 0.00029992803492742166,
|
| 1262 |
+
"loss": 0.2678,
|
| 1263 |
+
"step": 8950
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.29853716787740076,
|
| 1267 |
+
"grad_norm": 0.37581753730773926,
|
| 1268 |
+
"learning_rate": 0.0002999228484569734,
|
| 1269 |
+
"loss": 0.258,
|
| 1270 |
+
"step": 9000
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.3001957076989418,
|
| 1274 |
+
"grad_norm": 0.36842867732048035,
|
| 1275 |
+
"learning_rate": 0.0002999174815877307,
|
| 1276 |
+
"loss": 0.241,
|
| 1277 |
+
"step": 9050
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.30185424752048295,
|
| 1281 |
+
"grad_norm": 0.36468687653541565,
|
| 1282 |
+
"learning_rate": 0.0002999119343261515,
|
| 1283 |
+
"loss": 0.2601,
|
| 1284 |
+
"step": 9100
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.3035127873420241,
|
| 1288 |
+
"grad_norm": 0.44215506315231323,
|
| 1289 |
+
"learning_rate": 0.0002999062066789106,
|
| 1290 |
+
"loss": 0.2577,
|
| 1291 |
+
"step": 9150
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.3051713271635652,
|
| 1295 |
+
"grad_norm": 0.379226952791214,
|
| 1296 |
+
"learning_rate": 0.00029990029865290007,
|
| 1297 |
+
"loss": 0.2509,
|
| 1298 |
+
"step": 9200
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.30682986698510634,
|
| 1302 |
+
"grad_norm": 0.3723479211330414,
|
| 1303 |
+
"learning_rate": 0.0002998942102552288,
|
| 1304 |
+
"loss": 0.242,
|
| 1305 |
+
"step": 9250
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.3084884068066474,
|
| 1309 |
+
"grad_norm": 0.34578976035118103,
|
| 1310 |
+
"learning_rate": 0.00029988794149322283,
|
| 1311 |
+
"loss": 0.2791,
|
| 1312 |
+
"step": 9300
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.31014694662818854,
|
| 1316 |
+
"grad_norm": 0.33455216884613037,
|
| 1317 |
+
"learning_rate": 0.0002998814923744253,
|
| 1318 |
+
"loss": 0.2618,
|
| 1319 |
+
"step": 9350
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.31180548644972966,
|
| 1323 |
+
"grad_norm": 0.3505556583404541,
|
| 1324 |
+
"learning_rate": 0.0002998748629065962,
|
| 1325 |
+
"loss": 0.2416,
|
| 1326 |
+
"step": 9400
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.3134640262712708,
|
| 1330 |
+
"grad_norm": 0.3335639536380768,
|
| 1331 |
+
"learning_rate": 0.00029986805309771276,
|
| 1332 |
+
"loss": 0.2321,
|
| 1333 |
+
"step": 9450
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.31512256609281186,
|
| 1337 |
+
"grad_norm": 0.31904950737953186,
|
| 1338 |
+
"learning_rate": 0.00029986106295596884,
|
| 1339 |
+
"loss": 0.2405,
|
| 1340 |
+
"step": 9500
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.316781105914353,
|
| 1344 |
+
"grad_norm": 0.41776758432388306,
|
| 1345 |
+
"learning_rate": 0.0002998538924897757,
|
| 1346 |
+
"loss": 0.2359,
|
| 1347 |
+
"step": 9550
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.3184396457358941,
|
| 1351 |
+
"grad_norm": 0.3445201814174652,
|
| 1352 |
+
"learning_rate": 0.0002998465417077613,
|
| 1353 |
+
"loss": 0.2451,
|
| 1354 |
+
"step": 9600
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.32009818555743524,
|
| 1358 |
+
"grad_norm": 0.36852407455444336,
|
| 1359 |
+
"learning_rate": 0.0002998390106187708,
|
| 1360 |
+
"loss": 0.238,
|
| 1361 |
+
"step": 9650
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.32175672537897637,
|
| 1365 |
+
"grad_norm": 0.3637458384037018,
|
| 1366 |
+
"learning_rate": 0.00029983129923186614,
|
| 1367 |
+
"loss": 0.2286,
|
| 1368 |
+
"step": 9700
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.32341526520051744,
|
| 1372 |
+
"grad_norm": 0.35523203015327454,
|
| 1373 |
+
"learning_rate": 0.00029982340755632615,
|
| 1374 |
+
"loss": 0.2649,
|
| 1375 |
+
"step": 9750
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.32507380502205857,
|
| 1379 |
+
"grad_norm": 0.4552542567253113,
|
| 1380 |
+
"learning_rate": 0.00029981533560164683,
|
| 1381 |
+
"loss": 0.2462,
|
| 1382 |
+
"step": 9800
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.3267323448435997,
|
| 1386 |
+
"grad_norm": 0.3182304799556732,
|
| 1387 |
+
"learning_rate": 0.0002998070833775409,
|
| 1388 |
+
"loss": 0.259,
|
| 1389 |
+
"step": 9850
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.3283908846651408,
|
| 1393 |
+
"grad_norm": 0.3835619390010834,
|
| 1394 |
+
"learning_rate": 0.0002997986508939382,
|
| 1395 |
+
"loss": 0.2628,
|
| 1396 |
+
"step": 9900
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.33004942448668195,
|
| 1400 |
+
"grad_norm": 0.3364975154399872,
|
| 1401 |
+
"learning_rate": 0.00029979003816098514,
|
| 1402 |
+
"loss": 0.2512,
|
| 1403 |
+
"step": 9950
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.331707964308223,
|
| 1407 |
+
"grad_norm": 0.3546120822429657,
|
| 1408 |
+
"learning_rate": 0.0002997812451890454,
|
| 1409 |
+
"loss": 0.2312,
|
| 1410 |
+
"step": 10000
|
| 1411 |
+
}
|
| 1412 |
+
],
|
| 1413 |
+
"logging_steps": 50,
|
| 1414 |
+
"max_steps": 150735,
|
| 1415 |
+
"num_input_tokens_seen": 0,
|
| 1416 |
+
"num_train_epochs": 5,
|
| 1417 |
+
"save_steps": 10000,
|
| 1418 |
+
"stateful_callbacks": {
|
| 1419 |
+
"TrainerControl": {
|
| 1420 |
+
"args": {
|
| 1421 |
+
"should_epoch_stop": false,
|
| 1422 |
+
"should_evaluate": false,
|
| 1423 |
+
"should_log": false,
|
| 1424 |
+
"should_save": true,
|
| 1425 |
+
"should_training_stop": false
|
| 1426 |
+
},
|
| 1427 |
+
"attributes": {}
|
| 1428 |
+
}
|
| 1429 |
+
},
|
| 1430 |
+
"total_flos": 0.0,
|
| 1431 |
+
"train_batch_size": 16,
|
| 1432 |
+
"trial_name": null,
|
| 1433 |
+
"trial_params": null
|
| 1434 |
+
}
|