Muhammad Khalifa
commited on
Commit
·
bb2115a
1
Parent(s):
5545560
first commit
Browse files- added_tokens.json +6 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +109 -0
- spiece.model +3 -0
- tokenizer_config.json +112 -0
- trainer_state.json +1168 -0
- training_args.bin +3 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<<": 32100,
|
| 3 |
+
">>": 32101,
|
| 4 |
+
"[CLS]": 32102,
|
| 5 |
+
"[SEP]": 32103
|
| 6 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6091396de00ec751d44a1824c300f38d18610001cce6cd4d774028aec5d0d57a
|
| 3 |
+
size 2738190469
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6f2f69b62a083313a0e8aab1ff512204e74b59f4ec05bd1b8fcb6748136fac06
|
| 3 |
+
size 3136877397
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c13ec95721adb5a43cf2ead2491d5783907d164b3c421db1ad1b6780c1aace39
|
| 3 |
+
size 14575
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:15457ff710dd30841f265008d568d4bf44cc1bf70cd650bffc4b67bc9ba40a88
|
| 3 |
+
size 627
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,109 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<extra_id_0>",
|
| 4 |
+
"<extra_id_1>",
|
| 5 |
+
"<extra_id_2>",
|
| 6 |
+
"<extra_id_3>",
|
| 7 |
+
"<extra_id_4>",
|
| 8 |
+
"<extra_id_5>",
|
| 9 |
+
"<extra_id_6>",
|
| 10 |
+
"<extra_id_7>",
|
| 11 |
+
"<extra_id_8>",
|
| 12 |
+
"<extra_id_9>",
|
| 13 |
+
"<extra_id_10>",
|
| 14 |
+
"<extra_id_11>",
|
| 15 |
+
"<extra_id_12>",
|
| 16 |
+
"<extra_id_13>",
|
| 17 |
+
"<extra_id_14>",
|
| 18 |
+
"<extra_id_15>",
|
| 19 |
+
"<extra_id_16>",
|
| 20 |
+
"<extra_id_17>",
|
| 21 |
+
"<extra_id_18>",
|
| 22 |
+
"<extra_id_19>",
|
| 23 |
+
"<extra_id_20>",
|
| 24 |
+
"<extra_id_21>",
|
| 25 |
+
"<extra_id_22>",
|
| 26 |
+
"<extra_id_23>",
|
| 27 |
+
"<extra_id_24>",
|
| 28 |
+
"<extra_id_25>",
|
| 29 |
+
"<extra_id_26>",
|
| 30 |
+
"<extra_id_27>",
|
| 31 |
+
"<extra_id_28>",
|
| 32 |
+
"<extra_id_29>",
|
| 33 |
+
"<extra_id_30>",
|
| 34 |
+
"<extra_id_31>",
|
| 35 |
+
"<extra_id_32>",
|
| 36 |
+
"<extra_id_33>",
|
| 37 |
+
"<extra_id_34>",
|
| 38 |
+
"<extra_id_35>",
|
| 39 |
+
"<extra_id_36>",
|
| 40 |
+
"<extra_id_37>",
|
| 41 |
+
"<extra_id_38>",
|
| 42 |
+
"<extra_id_39>",
|
| 43 |
+
"<extra_id_40>",
|
| 44 |
+
"<extra_id_41>",
|
| 45 |
+
"<extra_id_42>",
|
| 46 |
+
"<extra_id_43>",
|
| 47 |
+
"<extra_id_44>",
|
| 48 |
+
"<extra_id_45>",
|
| 49 |
+
"<extra_id_46>",
|
| 50 |
+
"<extra_id_47>",
|
| 51 |
+
"<extra_id_48>",
|
| 52 |
+
"<extra_id_49>",
|
| 53 |
+
"<extra_id_50>",
|
| 54 |
+
"<extra_id_51>",
|
| 55 |
+
"<extra_id_52>",
|
| 56 |
+
"<extra_id_53>",
|
| 57 |
+
"<extra_id_54>",
|
| 58 |
+
"<extra_id_55>",
|
| 59 |
+
"<extra_id_56>",
|
| 60 |
+
"<extra_id_57>",
|
| 61 |
+
"<extra_id_58>",
|
| 62 |
+
"<extra_id_59>",
|
| 63 |
+
"<extra_id_60>",
|
| 64 |
+
"<extra_id_61>",
|
| 65 |
+
"<extra_id_62>",
|
| 66 |
+
"<extra_id_63>",
|
| 67 |
+
"<extra_id_64>",
|
| 68 |
+
"<extra_id_65>",
|
| 69 |
+
"<extra_id_66>",
|
| 70 |
+
"<extra_id_67>",
|
| 71 |
+
"<extra_id_68>",
|
| 72 |
+
"<extra_id_69>",
|
| 73 |
+
"<extra_id_70>",
|
| 74 |
+
"<extra_id_71>",
|
| 75 |
+
"<extra_id_72>",
|
| 76 |
+
"<extra_id_73>",
|
| 77 |
+
"<extra_id_74>",
|
| 78 |
+
"<extra_id_75>",
|
| 79 |
+
"<extra_id_76>",
|
| 80 |
+
"<extra_id_77>",
|
| 81 |
+
"<extra_id_78>",
|
| 82 |
+
"<extra_id_79>",
|
| 83 |
+
"<extra_id_80>",
|
| 84 |
+
"<extra_id_81>",
|
| 85 |
+
"<extra_id_82>",
|
| 86 |
+
"<extra_id_83>",
|
| 87 |
+
"<extra_id_84>",
|
| 88 |
+
"<extra_id_85>",
|
| 89 |
+
"<extra_id_86>",
|
| 90 |
+
"<extra_id_87>",
|
| 91 |
+
"<extra_id_88>",
|
| 92 |
+
"<extra_id_89>",
|
| 93 |
+
"<extra_id_90>",
|
| 94 |
+
"<extra_id_91>",
|
| 95 |
+
"<extra_id_92>",
|
| 96 |
+
"<extra_id_93>",
|
| 97 |
+
"<extra_id_94>",
|
| 98 |
+
"<extra_id_95>",
|
| 99 |
+
"<extra_id_96>",
|
| 100 |
+
"<extra_id_97>",
|
| 101 |
+
"<extra_id_98>",
|
| 102 |
+
"<extra_id_99>"
|
| 103 |
+
],
|
| 104 |
+
"cls_token": "[CLS]",
|
| 105 |
+
"eos_token": "</s>",
|
| 106 |
+
"pad_token": "<pad>",
|
| 107 |
+
"sep_token": "[SEP]",
|
| 108 |
+
"unk_token": "<unk>"
|
| 109 |
+
}
|
spiece.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
|
| 3 |
+
size 791656
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,112 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<extra_id_0>",
|
| 4 |
+
"<extra_id_1>",
|
| 5 |
+
"<extra_id_2>",
|
| 6 |
+
"<extra_id_3>",
|
| 7 |
+
"<extra_id_4>",
|
| 8 |
+
"<extra_id_5>",
|
| 9 |
+
"<extra_id_6>",
|
| 10 |
+
"<extra_id_7>",
|
| 11 |
+
"<extra_id_8>",
|
| 12 |
+
"<extra_id_9>",
|
| 13 |
+
"<extra_id_10>",
|
| 14 |
+
"<extra_id_11>",
|
| 15 |
+
"<extra_id_12>",
|
| 16 |
+
"<extra_id_13>",
|
| 17 |
+
"<extra_id_14>",
|
| 18 |
+
"<extra_id_15>",
|
| 19 |
+
"<extra_id_16>",
|
| 20 |
+
"<extra_id_17>",
|
| 21 |
+
"<extra_id_18>",
|
| 22 |
+
"<extra_id_19>",
|
| 23 |
+
"<extra_id_20>",
|
| 24 |
+
"<extra_id_21>",
|
| 25 |
+
"<extra_id_22>",
|
| 26 |
+
"<extra_id_23>",
|
| 27 |
+
"<extra_id_24>",
|
| 28 |
+
"<extra_id_25>",
|
| 29 |
+
"<extra_id_26>",
|
| 30 |
+
"<extra_id_27>",
|
| 31 |
+
"<extra_id_28>",
|
| 32 |
+
"<extra_id_29>",
|
| 33 |
+
"<extra_id_30>",
|
| 34 |
+
"<extra_id_31>",
|
| 35 |
+
"<extra_id_32>",
|
| 36 |
+
"<extra_id_33>",
|
| 37 |
+
"<extra_id_34>",
|
| 38 |
+
"<extra_id_35>",
|
| 39 |
+
"<extra_id_36>",
|
| 40 |
+
"<extra_id_37>",
|
| 41 |
+
"<extra_id_38>",
|
| 42 |
+
"<extra_id_39>",
|
| 43 |
+
"<extra_id_40>",
|
| 44 |
+
"<extra_id_41>",
|
| 45 |
+
"<extra_id_42>",
|
| 46 |
+
"<extra_id_43>",
|
| 47 |
+
"<extra_id_44>",
|
| 48 |
+
"<extra_id_45>",
|
| 49 |
+
"<extra_id_46>",
|
| 50 |
+
"<extra_id_47>",
|
| 51 |
+
"<extra_id_48>",
|
| 52 |
+
"<extra_id_49>",
|
| 53 |
+
"<extra_id_50>",
|
| 54 |
+
"<extra_id_51>",
|
| 55 |
+
"<extra_id_52>",
|
| 56 |
+
"<extra_id_53>",
|
| 57 |
+
"<extra_id_54>",
|
| 58 |
+
"<extra_id_55>",
|
| 59 |
+
"<extra_id_56>",
|
| 60 |
+
"<extra_id_57>",
|
| 61 |
+
"<extra_id_58>",
|
| 62 |
+
"<extra_id_59>",
|
| 63 |
+
"<extra_id_60>",
|
| 64 |
+
"<extra_id_61>",
|
| 65 |
+
"<extra_id_62>",
|
| 66 |
+
"<extra_id_63>",
|
| 67 |
+
"<extra_id_64>",
|
| 68 |
+
"<extra_id_65>",
|
| 69 |
+
"<extra_id_66>",
|
| 70 |
+
"<extra_id_67>",
|
| 71 |
+
"<extra_id_68>",
|
| 72 |
+
"<extra_id_69>",
|
| 73 |
+
"<extra_id_70>",
|
| 74 |
+
"<extra_id_71>",
|
| 75 |
+
"<extra_id_72>",
|
| 76 |
+
"<extra_id_73>",
|
| 77 |
+
"<extra_id_74>",
|
| 78 |
+
"<extra_id_75>",
|
| 79 |
+
"<extra_id_76>",
|
| 80 |
+
"<extra_id_77>",
|
| 81 |
+
"<extra_id_78>",
|
| 82 |
+
"<extra_id_79>",
|
| 83 |
+
"<extra_id_80>",
|
| 84 |
+
"<extra_id_81>",
|
| 85 |
+
"<extra_id_82>",
|
| 86 |
+
"<extra_id_83>",
|
| 87 |
+
"<extra_id_84>",
|
| 88 |
+
"<extra_id_85>",
|
| 89 |
+
"<extra_id_86>",
|
| 90 |
+
"<extra_id_87>",
|
| 91 |
+
"<extra_id_88>",
|
| 92 |
+
"<extra_id_89>",
|
| 93 |
+
"<extra_id_90>",
|
| 94 |
+
"<extra_id_91>",
|
| 95 |
+
"<extra_id_92>",
|
| 96 |
+
"<extra_id_93>",
|
| 97 |
+
"<extra_id_94>",
|
| 98 |
+
"<extra_id_95>",
|
| 99 |
+
"<extra_id_96>",
|
| 100 |
+
"<extra_id_97>",
|
| 101 |
+
"<extra_id_98>",
|
| 102 |
+
"<extra_id_99>"
|
| 103 |
+
],
|
| 104 |
+
"clean_up_tokenization_spaces": true,
|
| 105 |
+
"eos_token": "</s>",
|
| 106 |
+
"extra_ids": 100,
|
| 107 |
+
"model_max_length": 512,
|
| 108 |
+
"pad_token": "<pad>",
|
| 109 |
+
"sp_model_kwargs": {},
|
| 110 |
+
"tokenizer_class": "T5Tokenizer",
|
| 111 |
+
"unk_token": "<unk>"
|
| 112 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1168 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.5993568897247314,
|
| 3 |
+
"best_model_checkpoint": "ckpts/discrim/contrastive_models/gen.t5.large/flan_t5_large_tso/checkpoint-7200",
|
| 4 |
+
"epoch": 1.9380888290713325,
|
| 5 |
+
"global_step": 7200,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 2.6905829596412554e-06,
|
| 13 |
+
"loss": 0.9993,
|
| 14 |
+
"step": 50
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.03,
|
| 18 |
+
"learning_rate": 5.381165919282511e-06,
|
| 19 |
+
"loss": 0.9989,
|
| 20 |
+
"step": 100
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.04,
|
| 24 |
+
"learning_rate": 8.071748878923767e-06,
|
| 25 |
+
"loss": 0.9961,
|
| 26 |
+
"step": 150
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.05,
|
| 30 |
+
"learning_rate": 1.0762331838565022e-05,
|
| 31 |
+
"loss": 0.986,
|
| 32 |
+
"step": 200
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.05,
|
| 36 |
+
"eval_loss": 0.979211688041687,
|
| 37 |
+
"eval_runtime": 304.3772,
|
| 38 |
+
"eval_samples_per_second": 77.296,
|
| 39 |
+
"eval_steps_per_second": 9.662,
|
| 40 |
+
"step": 200
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.07,
|
| 44 |
+
"learning_rate": 1.345291479820628e-05,
|
| 45 |
+
"loss": 0.949,
|
| 46 |
+
"step": 250
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.08,
|
| 50 |
+
"learning_rate": 1.6143497757847534e-05,
|
| 51 |
+
"loss": 0.8611,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.09,
|
| 56 |
+
"learning_rate": 1.883408071748879e-05,
|
| 57 |
+
"loss": 0.7761,
|
| 58 |
+
"step": 350
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.11,
|
| 62 |
+
"learning_rate": 2.1524663677130043e-05,
|
| 63 |
+
"loss": 0.7103,
|
| 64 |
+
"step": 400
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.11,
|
| 68 |
+
"eval_loss": 0.73173987865448,
|
| 69 |
+
"eval_runtime": 304.6199,
|
| 70 |
+
"eval_samples_per_second": 77.234,
|
| 71 |
+
"eval_steps_per_second": 9.655,
|
| 72 |
+
"step": 400
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.12,
|
| 76 |
+
"learning_rate": 2.42152466367713e-05,
|
| 77 |
+
"loss": 0.6447,
|
| 78 |
+
"step": 450
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.13,
|
| 82 |
+
"learning_rate": 2.690582959641256e-05,
|
| 83 |
+
"loss": 0.6239,
|
| 84 |
+
"step": 500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.15,
|
| 88 |
+
"learning_rate": 2.9596412556053812e-05,
|
| 89 |
+
"loss": 0.6433,
|
| 90 |
+
"step": 550
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.16,
|
| 94 |
+
"learning_rate": 3.228699551569507e-05,
|
| 95 |
+
"loss": 0.5925,
|
| 96 |
+
"step": 600
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.16,
|
| 100 |
+
"eval_loss": 0.7419408559799194,
|
| 101 |
+
"eval_runtime": 305.2185,
|
| 102 |
+
"eval_samples_per_second": 77.082,
|
| 103 |
+
"eval_steps_per_second": 9.636,
|
| 104 |
+
"step": 600
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.17,
|
| 108 |
+
"learning_rate": 3.4977578475336325e-05,
|
| 109 |
+
"loss": 0.5665,
|
| 110 |
+
"step": 650
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.19,
|
| 114 |
+
"learning_rate": 3.766816143497758e-05,
|
| 115 |
+
"loss": 0.5167,
|
| 116 |
+
"step": 700
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.2,
|
| 120 |
+
"learning_rate": 4.035874439461884e-05,
|
| 121 |
+
"loss": 0.5378,
|
| 122 |
+
"step": 750
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.22,
|
| 126 |
+
"learning_rate": 4.304932735426009e-05,
|
| 127 |
+
"loss": 0.4876,
|
| 128 |
+
"step": 800
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.22,
|
| 132 |
+
"eval_loss": 0.7308951020240784,
|
| 133 |
+
"eval_runtime": 305.4422,
|
| 134 |
+
"eval_samples_per_second": 77.026,
|
| 135 |
+
"eval_steps_per_second": 9.629,
|
| 136 |
+
"step": 800
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.23,
|
| 140 |
+
"learning_rate": 4.573991031390134e-05,
|
| 141 |
+
"loss": 0.4874,
|
| 142 |
+
"step": 850
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.24,
|
| 146 |
+
"learning_rate": 4.84304932735426e-05,
|
| 147 |
+
"loss": 0.4476,
|
| 148 |
+
"step": 900
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.26,
|
| 152 |
+
"learning_rate": 5.1121076233183856e-05,
|
| 153 |
+
"loss": 0.4695,
|
| 154 |
+
"step": 950
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.27,
|
| 158 |
+
"learning_rate": 5.381165919282512e-05,
|
| 159 |
+
"loss": 0.4334,
|
| 160 |
+
"step": 1000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.27,
|
| 164 |
+
"eval_loss": 0.6671885848045349,
|
| 165 |
+
"eval_runtime": 305.4747,
|
| 166 |
+
"eval_samples_per_second": 77.018,
|
| 167 |
+
"eval_steps_per_second": 9.628,
|
| 168 |
+
"step": 1000
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 0.28,
|
| 172 |
+
"learning_rate": 5.6502242152466375e-05,
|
| 173 |
+
"loss": 0.389,
|
| 174 |
+
"step": 1050
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.3,
|
| 178 |
+
"learning_rate": 5.9192825112107625e-05,
|
| 179 |
+
"loss": 0.3758,
|
| 180 |
+
"step": 1100
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 0.31,
|
| 184 |
+
"learning_rate": 5.999940510930071e-05,
|
| 185 |
+
"loss": 0.3737,
|
| 186 |
+
"step": 1150
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.32,
|
| 190 |
+
"learning_rate": 5.999649141573284e-05,
|
| 191 |
+
"loss": 0.3586,
|
| 192 |
+
"step": 1200
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.32,
|
| 196 |
+
"eval_loss": 0.6958568096160889,
|
| 197 |
+
"eval_runtime": 305.3709,
|
| 198 |
+
"eval_samples_per_second": 77.044,
|
| 199 |
+
"eval_steps_per_second": 9.631,
|
| 200 |
+
"step": 1200
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.34,
|
| 204 |
+
"learning_rate": 5.9991149889189245e-05,
|
| 205 |
+
"loss": 0.3154,
|
| 206 |
+
"step": 1250
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.35,
|
| 210 |
+
"learning_rate": 5.9983380961998294e-05,
|
| 211 |
+
"loss": 0.3202,
|
| 212 |
+
"step": 1300
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.36,
|
| 216 |
+
"learning_rate": 5.997318526295544e-05,
|
| 217 |
+
"loss": 0.3232,
|
| 218 |
+
"step": 1350
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.38,
|
| 222 |
+
"learning_rate": 5.996056361727235e-05,
|
| 223 |
+
"loss": 0.3016,
|
| 224 |
+
"step": 1400
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.38,
|
| 228 |
+
"eval_loss": 0.6629464030265808,
|
| 229 |
+
"eval_runtime": 305.4391,
|
| 230 |
+
"eval_samples_per_second": 77.027,
|
| 231 |
+
"eval_steps_per_second": 9.629,
|
| 232 |
+
"step": 1400
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.39,
|
| 236 |
+
"learning_rate": 5.9945517046510075e-05,
|
| 237 |
+
"loss": 0.2673,
|
| 238 |
+
"step": 1450
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.4,
|
| 242 |
+
"learning_rate": 5.992804676849642e-05,
|
| 243 |
+
"loss": 0.2558,
|
| 244 |
+
"step": 1500
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 0.42,
|
| 248 |
+
"learning_rate": 5.990815419722729e-05,
|
| 249 |
+
"loss": 0.2566,
|
| 250 |
+
"step": 1550
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.43,
|
| 254 |
+
"learning_rate": 5.988584094275237e-05,
|
| 255 |
+
"loss": 0.2496,
|
| 256 |
+
"step": 1600
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.43,
|
| 260 |
+
"eval_loss": 0.6636276841163635,
|
| 261 |
+
"eval_runtime": 305.4888,
|
| 262 |
+
"eval_samples_per_second": 77.014,
|
| 263 |
+
"eval_steps_per_second": 9.627,
|
| 264 |
+
"step": 1600
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.44,
|
| 268 |
+
"learning_rate": 5.9861108811044676e-05,
|
| 269 |
+
"loss": 0.2273,
|
| 270 |
+
"step": 1650
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 0.46,
|
| 274 |
+
"learning_rate": 5.983395980385451e-05,
|
| 275 |
+
"loss": 0.2121,
|
| 276 |
+
"step": 1700
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.47,
|
| 280 |
+
"learning_rate": 5.9804396118547345e-05,
|
| 281 |
+
"loss": 0.1966,
|
| 282 |
+
"step": 1750
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.48,
|
| 286 |
+
"learning_rate": 5.977242014792602e-05,
|
| 287 |
+
"loss": 0.1929,
|
| 288 |
+
"step": 1800
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.48,
|
| 292 |
+
"eval_loss": 0.6448654532432556,
|
| 293 |
+
"eval_runtime": 305.5903,
|
| 294 |
+
"eval_samples_per_second": 76.989,
|
| 295 |
+
"eval_steps_per_second": 9.624,
|
| 296 |
+
"step": 1800
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.5,
|
| 300 |
+
"learning_rate": 5.9738034480037094e-05,
|
| 301 |
+
"loss": 0.1797,
|
| 302 |
+
"step": 1850
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.51,
|
| 306 |
+
"learning_rate": 5.970124189796133e-05,
|
| 307 |
+
"loss": 0.1807,
|
| 308 |
+
"step": 1900
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.52,
|
| 312 |
+
"learning_rate": 5.966204537958846e-05,
|
| 313 |
+
"loss": 0.1438,
|
| 314 |
+
"step": 1950
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.54,
|
| 318 |
+
"learning_rate": 5.962044809737618e-05,
|
| 319 |
+
"loss": 0.1721,
|
| 320 |
+
"step": 2000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.54,
|
| 324 |
+
"eval_loss": 0.6665562391281128,
|
| 325 |
+
"eval_runtime": 305.4858,
|
| 326 |
+
"eval_samples_per_second": 77.015,
|
| 327 |
+
"eval_steps_per_second": 9.627,
|
| 328 |
+
"step": 2000
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.55,
|
| 332 |
+
"learning_rate": 5.957645341809336e-05,
|
| 333 |
+
"loss": 0.1555,
|
| 334 |
+
"step": 2050
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 0.57,
|
| 338 |
+
"learning_rate": 5.953006490254755e-05,
|
| 339 |
+
"loss": 0.1543,
|
| 340 |
+
"step": 2100
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 0.58,
|
| 344 |
+
"learning_rate": 5.9481286305296774e-05,
|
| 345 |
+
"loss": 0.1498,
|
| 346 |
+
"step": 2150
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.59,
|
| 350 |
+
"learning_rate": 5.943012157434567e-05,
|
| 351 |
+
"loss": 0.1495,
|
| 352 |
+
"step": 2200
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.59,
|
| 356 |
+
"eval_loss": 0.6751571893692017,
|
| 357 |
+
"eval_runtime": 305.7212,
|
| 358 |
+
"eval_samples_per_second": 76.956,
|
| 359 |
+
"eval_steps_per_second": 9.62,
|
| 360 |
+
"step": 2200
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.61,
|
| 364 |
+
"learning_rate": 5.9376574850825905e-05,
|
| 365 |
+
"loss": 0.1206,
|
| 366 |
+
"step": 2250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.62,
|
| 370 |
+
"learning_rate": 5.9320650468661066e-05,
|
| 371 |
+
"loss": 0.1261,
|
| 372 |
+
"step": 2300
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 0.63,
|
| 376 |
+
"learning_rate": 5.926235295421582e-05,
|
| 377 |
+
"loss": 0.1173,
|
| 378 |
+
"step": 2350
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 0.65,
|
| 382 |
+
"learning_rate": 5.92016870259296e-05,
|
| 383 |
+
"loss": 0.1132,
|
| 384 |
+
"step": 2400
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.65,
|
| 388 |
+
"eval_loss": 0.7012338638305664,
|
| 389 |
+
"eval_runtime": 305.6366,
|
| 390 |
+
"eval_samples_per_second": 76.977,
|
| 391 |
+
"eval_steps_per_second": 9.623,
|
| 392 |
+
"step": 2400
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.66,
|
| 396 |
+
"learning_rate": 5.9138657593934704e-05,
|
| 397 |
+
"loss": 0.1225,
|
| 398 |
+
"step": 2450
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.67,
|
| 402 |
+
"learning_rate": 5.907326975965887e-05,
|
| 403 |
+
"loss": 0.112,
|
| 404 |
+
"step": 2500
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.69,
|
| 408 |
+
"learning_rate": 5.9005528815412394e-05,
|
| 409 |
+
"loss": 0.1105,
|
| 410 |
+
"step": 2550
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.7,
|
| 414 |
+
"learning_rate": 5.893544024395977e-05,
|
| 415 |
+
"loss": 0.1058,
|
| 416 |
+
"step": 2600
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.7,
|
| 420 |
+
"eval_loss": 0.6904404163360596,
|
| 421 |
+
"eval_runtime": 305.6312,
|
| 422 |
+
"eval_samples_per_second": 76.978,
|
| 423 |
+
"eval_steps_per_second": 9.623,
|
| 424 |
+
"step": 2600
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.71,
|
| 428 |
+
"learning_rate": 5.886300971807594e-05,
|
| 429 |
+
"loss": 0.0855,
|
| 430 |
+
"step": 2650
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.73,
|
| 434 |
+
"learning_rate": 5.878824310008718e-05,
|
| 435 |
+
"loss": 0.088,
|
| 436 |
+
"step": 2700
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.74,
|
| 440 |
+
"learning_rate": 5.871114644139656e-05,
|
| 441 |
+
"loss": 0.0926,
|
| 442 |
+
"step": 2750
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 0.75,
|
| 446 |
+
"learning_rate": 5.86317259819942e-05,
|
| 447 |
+
"loss": 0.1057,
|
| 448 |
+
"step": 2800
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 0.75,
|
| 452 |
+
"eval_loss": 0.6426856517791748,
|
| 453 |
+
"eval_runtime": 305.7094,
|
| 454 |
+
"eval_samples_per_second": 76.959,
|
| 455 |
+
"eval_steps_per_second": 9.62,
|
| 456 |
+
"step": 2800
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 0.77,
|
| 460 |
+
"learning_rate": 5.854998814995223e-05,
|
| 461 |
+
"loss": 0.0846,
|
| 462 |
+
"step": 2850
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 0.78,
|
| 466 |
+
"learning_rate": 5.846593956090451e-05,
|
| 467 |
+
"loss": 0.0838,
|
| 468 |
+
"step": 2900
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 0.79,
|
| 472 |
+
"learning_rate": 5.837958701751115e-05,
|
| 473 |
+
"loss": 0.0827,
|
| 474 |
+
"step": 2950
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.81,
|
| 478 |
+
"learning_rate": 5.8290937508907975e-05,
|
| 479 |
+
"loss": 0.0818,
|
| 480 |
+
"step": 3000
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.81,
|
| 484 |
+
"eval_loss": 0.6715692281723022,
|
| 485 |
+
"eval_runtime": 305.6575,
|
| 486 |
+
"eval_samples_per_second": 76.972,
|
| 487 |
+
"eval_steps_per_second": 9.622,
|
| 488 |
+
"step": 3000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.82,
|
| 492 |
+
"learning_rate": 5.819999821014079e-05,
|
| 493 |
+
"loss": 0.114,
|
| 494 |
+
"step": 3050
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.83,
|
| 498 |
+
"learning_rate": 5.81067764815847e-05,
|
| 499 |
+
"loss": 0.0742,
|
| 500 |
+
"step": 3100
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.85,
|
| 504 |
+
"learning_rate": 5.801127986834834e-05,
|
| 505 |
+
"loss": 0.0764,
|
| 506 |
+
"step": 3150
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.86,
|
| 510 |
+
"learning_rate": 5.791351609966323e-05,
|
| 511 |
+
"loss": 0.0893,
|
| 512 |
+
"step": 3200
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.86,
|
| 516 |
+
"eval_loss": 0.6607950329780579,
|
| 517 |
+
"eval_runtime": 305.6387,
|
| 518 |
+
"eval_samples_per_second": 76.977,
|
| 519 |
+
"eval_steps_per_second": 9.622,
|
| 520 |
+
"step": 3200
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.87,
|
| 524 |
+
"learning_rate": 5.781349308825816e-05,
|
| 525 |
+
"loss": 0.2025,
|
| 526 |
+
"step": 3250
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.89,
|
| 530 |
+
"learning_rate": 5.7711218929718786e-05,
|
| 531 |
+
"loss": 0.3151,
|
| 532 |
+
"step": 3300
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 0.9,
|
| 536 |
+
"learning_rate": 5.760670190183236e-05,
|
| 537 |
+
"loss": 0.3307,
|
| 538 |
+
"step": 3350
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 0.92,
|
| 542 |
+
"learning_rate": 5.74999504639178e-05,
|
| 543 |
+
"loss": 0.3306,
|
| 544 |
+
"step": 3400
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 0.92,
|
| 548 |
+
"eval_loss": 0.7396358251571655,
|
| 549 |
+
"eval_runtime": 305.7093,
|
| 550 |
+
"eval_samples_per_second": 76.959,
|
| 551 |
+
"eval_steps_per_second": 9.62,
|
| 552 |
+
"step": 3400
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 0.93,
|
| 556 |
+
"learning_rate": 5.739097325614097e-05,
|
| 557 |
+
"loss": 0.3255,
|
| 558 |
+
"step": 3450
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 0.94,
|
| 562 |
+
"learning_rate": 5.727977909881537e-05,
|
| 563 |
+
"loss": 0.317,
|
| 564 |
+
"step": 3500
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.96,
|
| 568 |
+
"learning_rate": 5.716637699168828e-05,
|
| 569 |
+
"loss": 0.3187,
|
| 570 |
+
"step": 3550
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.97,
|
| 574 |
+
"learning_rate": 5.705077611321233e-05,
|
| 575 |
+
"loss": 0.3105,
|
| 576 |
+
"step": 3600
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.97,
|
| 580 |
+
"eval_loss": 0.7177180051803589,
|
| 581 |
+
"eval_runtime": 305.5872,
|
| 582 |
+
"eval_samples_per_second": 76.989,
|
| 583 |
+
"eval_steps_per_second": 9.624,
|
| 584 |
+
"step": 3600
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.98,
|
| 588 |
+
"learning_rate": 5.69329858198026e-05,
|
| 589 |
+
"loss": 0.3129,
|
| 590 |
+
"step": 3650
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.0,
|
| 594 |
+
"learning_rate": 5.6813015645079355e-05,
|
| 595 |
+
"loss": 0.3237,
|
| 596 |
+
"step": 3700
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 1.01,
|
| 600 |
+
"learning_rate": 5.6690875299096415e-05,
|
| 601 |
+
"loss": 0.2951,
|
| 602 |
+
"step": 3750
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 1.02,
|
| 606 |
+
"learning_rate": 5.656657466755528e-05,
|
| 607 |
+
"loss": 0.2882,
|
| 608 |
+
"step": 3800
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 1.02,
|
| 612 |
+
"eval_loss": 0.7239081263542175,
|
| 613 |
+
"eval_runtime": 305.6494,
|
| 614 |
+
"eval_samples_per_second": 76.974,
|
| 615 |
+
"eval_steps_per_second": 9.622,
|
| 616 |
+
"step": 3800
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 1.04,
|
| 620 |
+
"learning_rate": 5.6440123811004955e-05,
|
| 621 |
+
"loss": 0.3036,
|
| 622 |
+
"step": 3850
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 1.05,
|
| 626 |
+
"learning_rate": 5.631153296402772e-05,
|
| 627 |
+
"loss": 0.2988,
|
| 628 |
+
"step": 3900
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 1.06,
|
| 632 |
+
"learning_rate": 5.6180812534410734e-05,
|
| 633 |
+
"loss": 0.2982,
|
| 634 |
+
"step": 3950
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 1.08,
|
| 638 |
+
"learning_rate": 5.604797310230373e-05,
|
| 639 |
+
"loss": 0.3034,
|
| 640 |
+
"step": 4000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.08,
|
| 644 |
+
"eval_loss": 0.7187855243682861,
|
| 645 |
+
"eval_runtime": 305.6191,
|
| 646 |
+
"eval_samples_per_second": 76.981,
|
| 647 |
+
"eval_steps_per_second": 9.623,
|
| 648 |
+
"step": 4000
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 1.09,
|
| 652 |
+
"learning_rate": 5.591302541936257e-05,
|
| 653 |
+
"loss": 0.3056,
|
| 654 |
+
"step": 4050
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.1,
|
| 658 |
+
"learning_rate": 5.577598040787914e-05,
|
| 659 |
+
"loss": 0.3106,
|
| 660 |
+
"step": 4100
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.12,
|
| 664 |
+
"learning_rate": 5.5636849159897306e-05,
|
| 665 |
+
"loss": 0.3083,
|
| 666 |
+
"step": 4150
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 1.13,
|
| 670 |
+
"learning_rate": 5.54956429363151e-05,
|
| 671 |
+
"loss": 0.2335,
|
| 672 |
+
"step": 4200
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 1.13,
|
| 676 |
+
"eval_loss": 0.6692826747894287,
|
| 677 |
+
"eval_runtime": 305.6153,
|
| 678 |
+
"eval_samples_per_second": 76.982,
|
| 679 |
+
"eval_steps_per_second": 9.623,
|
| 680 |
+
"step": 4200
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 1.14,
|
| 684 |
+
"learning_rate": 5.535237316597337e-05,
|
| 685 |
+
"loss": 0.0927,
|
| 686 |
+
"step": 4250
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 1.16,
|
| 690 |
+
"learning_rate": 5.5207051444730707e-05,
|
| 691 |
+
"loss": 0.0937,
|
| 692 |
+
"step": 4300
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 1.17,
|
| 696 |
+
"learning_rate": 5.505968953452494e-05,
|
| 697 |
+
"loss": 0.0998,
|
| 698 |
+
"step": 4350
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 1.18,
|
| 702 |
+
"learning_rate": 5.4910299362421125e-05,
|
| 703 |
+
"loss": 0.0881,
|
| 704 |
+
"step": 4400
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 1.18,
|
| 708 |
+
"eval_loss": 0.6463451385498047,
|
| 709 |
+
"eval_runtime": 305.7621,
|
| 710 |
+
"eval_samples_per_second": 76.945,
|
| 711 |
+
"eval_steps_per_second": 9.619,
|
| 712 |
+
"step": 4400
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 1.2,
|
| 716 |
+
"learning_rate": 5.4758893019646234e-05,
|
| 717 |
+
"loss": 0.0694,
|
| 718 |
+
"step": 4450
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 1.21,
|
| 722 |
+
"learning_rate": 5.460548276061054e-05,
|
| 723 |
+
"loss": 0.0806,
|
| 724 |
+
"step": 4500
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 1.22,
|
| 728 |
+
"learning_rate": 5.4450081001915694e-05,
|
| 729 |
+
"loss": 0.0735,
|
| 730 |
+
"step": 4550
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.24,
|
| 734 |
+
"learning_rate": 5.429270032134986e-05,
|
| 735 |
+
"loss": 0.0686,
|
| 736 |
+
"step": 4600
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 1.24,
|
| 740 |
+
"eval_loss": 0.7035936117172241,
|
| 741 |
+
"eval_runtime": 305.6859,
|
| 742 |
+
"eval_samples_per_second": 76.965,
|
| 743 |
+
"eval_steps_per_second": 9.621,
|
| 744 |
+
"step": 4600
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.25,
|
| 748 |
+
"learning_rate": 5.413335345686963e-05,
|
| 749 |
+
"loss": 0.0751,
|
| 750 |
+
"step": 4650
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 1.27,
|
| 754 |
+
"learning_rate": 5.397205330556912e-05,
|
| 755 |
+
"loss": 0.0794,
|
| 756 |
+
"step": 4700
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 1.28,
|
| 760 |
+
"learning_rate": 5.380881292263602e-05,
|
| 761 |
+
"loss": 0.0738,
|
| 762 |
+
"step": 4750
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 1.29,
|
| 766 |
+
"learning_rate": 5.364364552029504e-05,
|
| 767 |
+
"loss": 0.07,
|
| 768 |
+
"step": 4800
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 1.29,
|
| 772 |
+
"eval_loss": 0.6865878701210022,
|
| 773 |
+
"eval_runtime": 305.6253,
|
| 774 |
+
"eval_samples_per_second": 76.98,
|
| 775 |
+
"eval_steps_per_second": 9.623,
|
| 776 |
+
"step": 4800
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 1.31,
|
| 780 |
+
"learning_rate": 5.347656446673849e-05,
|
| 781 |
+
"loss": 0.0601,
|
| 782 |
+
"step": 4850
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 1.32,
|
| 786 |
+
"learning_rate": 5.330758328504432e-05,
|
| 787 |
+
"loss": 0.0622,
|
| 788 |
+
"step": 4900
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 1.33,
|
| 792 |
+
"learning_rate": 5.313671565208157e-05,
|
| 793 |
+
"loss": 0.0617,
|
| 794 |
+
"step": 4950
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 1.35,
|
| 798 |
+
"learning_rate": 5.296397539740344e-05,
|
| 799 |
+
"loss": 0.0655,
|
| 800 |
+
"step": 5000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.35,
|
| 804 |
+
"eval_loss": 0.6813921332359314,
|
| 805 |
+
"eval_runtime": 305.6628,
|
| 806 |
+
"eval_samples_per_second": 76.97,
|
| 807 |
+
"eval_steps_per_second": 9.622,
|
| 808 |
+
"step": 5000
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 1.36,
|
| 812 |
+
"learning_rate": 5.2789376502127956e-05,
|
| 813 |
+
"loss": 0.0621,
|
| 814 |
+
"step": 5050
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.37,
|
| 818 |
+
"learning_rate": 5.261293309780635e-05,
|
| 819 |
+
"loss": 0.0651,
|
| 820 |
+
"step": 5100
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 1.39,
|
| 824 |
+
"learning_rate": 5.243465946527933e-05,
|
| 825 |
+
"loss": 0.0529,
|
| 826 |
+
"step": 5150
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 1.4,
|
| 830 |
+
"learning_rate": 5.225457003352119e-05,
|
| 831 |
+
"loss": 0.0483,
|
| 832 |
+
"step": 5200
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 1.4,
|
| 836 |
+
"eval_loss": 0.6487281322479248,
|
| 837 |
+
"eval_runtime": 305.6821,
|
| 838 |
+
"eval_samples_per_second": 76.966,
|
| 839 |
+
"eval_steps_per_second": 9.621,
|
| 840 |
+
"step": 5200
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 1.41,
|
| 844 |
+
"learning_rate": 5.207267937847198e-05,
|
| 845 |
+
"loss": 0.0554,
|
| 846 |
+
"step": 5250
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 1.43,
|
| 850 |
+
"learning_rate": 5.188900222185781e-05,
|
| 851 |
+
"loss": 0.066,
|
| 852 |
+
"step": 5300
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 1.44,
|
| 856 |
+
"learning_rate": 5.170355342999926e-05,
|
| 857 |
+
"loss": 0.0502,
|
| 858 |
+
"step": 5350
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 1.45,
|
| 862 |
+
"learning_rate": 5.151634801260815e-05,
|
| 863 |
+
"loss": 0.0467,
|
| 864 |
+
"step": 5400
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.45,
|
| 868 |
+
"eval_loss": 0.6819436550140381,
|
| 869 |
+
"eval_runtime": 305.7363,
|
| 870 |
+
"eval_samples_per_second": 76.952,
|
| 871 |
+
"eval_steps_per_second": 9.619,
|
| 872 |
+
"step": 5400
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 1.47,
|
| 876 |
+
"learning_rate": 5.132740112157274e-05,
|
| 877 |
+
"loss": 0.0439,
|
| 878 |
+
"step": 5450
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.48,
|
| 882 |
+
"learning_rate": 5.1136728049731315e-05,
|
| 883 |
+
"loss": 0.0467,
|
| 884 |
+
"step": 5500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.49,
|
| 888 |
+
"learning_rate": 5.0944344229634474e-05,
|
| 889 |
+
"loss": 0.0454,
|
| 890 |
+
"step": 5550
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 1.51,
|
| 894 |
+
"learning_rate": 5.075026523229604e-05,
|
| 895 |
+
"loss": 0.0461,
|
| 896 |
+
"step": 5600
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 1.51,
|
| 900 |
+
"eval_loss": 0.6921408176422119,
|
| 901 |
+
"eval_runtime": 305.7277,
|
| 902 |
+
"eval_samples_per_second": 76.954,
|
| 903 |
+
"eval_steps_per_second": 9.62,
|
| 904 |
+
"step": 5600
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 1.52,
|
| 908 |
+
"learning_rate": 5.055450676593277e-05,
|
| 909 |
+
"loss": 0.0496,
|
| 910 |
+
"step": 5650
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 1.53,
|
| 914 |
+
"learning_rate": 5.0357084674693044e-05,
|
| 915 |
+
"loss": 0.0376,
|
| 916 |
+
"step": 5700
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 1.55,
|
| 920 |
+
"learning_rate": 5.015801493737437e-05,
|
| 921 |
+
"loss": 0.0465,
|
| 922 |
+
"step": 5750
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 1.56,
|
| 926 |
+
"learning_rate": 4.995731366613022e-05,
|
| 927 |
+
"loss": 0.0473,
|
| 928 |
+
"step": 5800
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 1.56,
|
| 932 |
+
"eval_loss": 0.6527886986732483,
|
| 933 |
+
"eval_runtime": 305.6041,
|
| 934 |
+
"eval_samples_per_second": 76.985,
|
| 935 |
+
"eval_steps_per_second": 9.624,
|
| 936 |
+
"step": 5800
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 1.57,
|
| 940 |
+
"learning_rate": 4.975499710516589e-05,
|
| 941 |
+
"loss": 0.0377,
|
| 942 |
+
"step": 5850
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 1.59,
|
| 946 |
+
"learning_rate": 4.9551081629423734e-05,
|
| 947 |
+
"loss": 0.0462,
|
| 948 |
+
"step": 5900
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.6,
|
| 952 |
+
"learning_rate": 4.934558374325787e-05,
|
| 953 |
+
"loss": 0.0378,
|
| 954 |
+
"step": 5950
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.62,
|
| 958 |
+
"learning_rate": 4.913852007909832e-05,
|
| 959 |
+
"loss": 0.0477,
|
| 960 |
+
"step": 6000
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 1.62,
|
| 964 |
+
"eval_loss": 0.6821478605270386,
|
| 965 |
+
"eval_runtime": 305.5702,
|
| 966 |
+
"eval_samples_per_second": 76.994,
|
| 967 |
+
"eval_steps_per_second": 9.625,
|
| 968 |
+
"step": 6000
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 1.63,
|
| 972 |
+
"learning_rate": 4.8929907396104834e-05,
|
| 973 |
+
"loss": 0.0443,
|
| 974 |
+
"step": 6050
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 1.64,
|
| 978 |
+
"learning_rate": 4.871976257881049e-05,
|
| 979 |
+
"loss": 0.0551,
|
| 980 |
+
"step": 6100
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 1.66,
|
| 984 |
+
"learning_rate": 4.850810263575504e-05,
|
| 985 |
+
"loss": 0.0518,
|
| 986 |
+
"step": 6150
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 1.67,
|
| 990 |
+
"learning_rate": 4.829494469810834e-05,
|
| 991 |
+
"loss": 0.0343,
|
| 992 |
+
"step": 6200
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 1.67,
|
| 996 |
+
"eval_loss": 0.6623004078865051,
|
| 997 |
+
"eval_runtime": 305.5922,
|
| 998 |
+
"eval_samples_per_second": 76.988,
|
| 999 |
+
"eval_steps_per_second": 9.624,
|
| 1000 |
+
"step": 6200
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 1.68,
|
| 1004 |
+
"learning_rate": 4.808030601828378e-05,
|
| 1005 |
+
"loss": 0.0461,
|
| 1006 |
+
"step": 6250
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 1.7,
|
| 1010 |
+
"learning_rate": 4.7864203968541914e-05,
|
| 1011 |
+
"loss": 0.0404,
|
| 1012 |
+
"step": 6300
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 1.71,
|
| 1016 |
+
"learning_rate": 4.7646656039584416e-05,
|
| 1017 |
+
"loss": 0.0303,
|
| 1018 |
+
"step": 6350
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.72,
|
| 1022 |
+
"learning_rate": 4.742767983913842e-05,
|
| 1023 |
+
"loss": 0.0376,
|
| 1024 |
+
"step": 6400
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 1.72,
|
| 1028 |
+
"eval_loss": 0.653499186038971,
|
| 1029 |
+
"eval_runtime": 305.6562,
|
| 1030 |
+
"eval_samples_per_second": 76.972,
|
| 1031 |
+
"eval_steps_per_second": 9.622,
|
| 1032 |
+
"step": 6400
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.74,
|
| 1036 |
+
"learning_rate": 4.720729309053138e-05,
|
| 1037 |
+
"loss": 0.0402,
|
| 1038 |
+
"step": 6450
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 1.75,
|
| 1042 |
+
"learning_rate": 4.698551363125665e-05,
|
| 1043 |
+
"loss": 0.0254,
|
| 1044 |
+
"step": 6500
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 1.76,
|
| 1048 |
+
"learning_rate": 4.67623594115297e-05,
|
| 1049 |
+
"loss": 0.0386,
|
| 1050 |
+
"step": 6550
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 1.78,
|
| 1054 |
+
"learning_rate": 4.6537848492835324e-05,
|
| 1055 |
+
"loss": 0.033,
|
| 1056 |
+
"step": 6600
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 1.78,
|
| 1060 |
+
"eval_loss": 0.696467399597168,
|
| 1061 |
+
"eval_runtime": 305.7738,
|
| 1062 |
+
"eval_samples_per_second": 76.943,
|
| 1063 |
+
"eval_steps_per_second": 9.618,
|
| 1064 |
+
"step": 6600
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 1.79,
|
| 1068 |
+
"learning_rate": 4.631199904646576e-05,
|
| 1069 |
+
"loss": 0.0262,
|
| 1070 |
+
"step": 6650
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 1.8,
|
| 1074 |
+
"learning_rate": 4.608482935204997e-05,
|
| 1075 |
+
"loss": 0.0298,
|
| 1076 |
+
"step": 6700
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 1.82,
|
| 1080 |
+
"learning_rate": 4.585635779607416e-05,
|
| 1081 |
+
"loss": 0.0288,
|
| 1082 |
+
"step": 6750
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 1.83,
|
| 1086 |
+
"learning_rate": 4.562660287039358e-05,
|
| 1087 |
+
"loss": 0.044,
|
| 1088 |
+
"step": 6800
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.83,
|
| 1092 |
+
"eval_loss": 0.6436501741409302,
|
| 1093 |
+
"eval_runtime": 305.7319,
|
| 1094 |
+
"eval_samples_per_second": 76.953,
|
| 1095 |
+
"eval_steps_per_second": 9.62,
|
| 1096 |
+
"step": 6800
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 1.84,
|
| 1100 |
+
"learning_rate": 4.539558317073588e-05,
|
| 1101 |
+
"loss": 0.0322,
|
| 1102 |
+
"step": 6850
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.86,
|
| 1106 |
+
"learning_rate": 4.516331739519603e-05,
|
| 1107 |
+
"loss": 0.0363,
|
| 1108 |
+
"step": 6900
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 1.87,
|
| 1112 |
+
"learning_rate": 4.492982434272291e-05,
|
| 1113 |
+
"loss": 0.0285,
|
| 1114 |
+
"step": 6950
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 1.88,
|
| 1118 |
+
"learning_rate": 4.469512291159781e-05,
|
| 1119 |
+
"loss": 0.0305,
|
| 1120 |
+
"step": 7000
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 1.88,
|
| 1124 |
+
"eval_loss": 0.6893681287765503,
|
| 1125 |
+
"eval_runtime": 305.7064,
|
| 1126 |
+
"eval_samples_per_second": 76.959,
|
| 1127 |
+
"eval_steps_per_second": 9.62,
|
| 1128 |
+
"step": 7000
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 1.9,
|
| 1132 |
+
"learning_rate": 4.4459232097904836e-05,
|
| 1133 |
+
"loss": 0.0292,
|
| 1134 |
+
"step": 7050
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 1.91,
|
| 1138 |
+
"learning_rate": 4.4222170993993444e-05,
|
| 1139 |
+
"loss": 0.0346,
|
| 1140 |
+
"step": 7100
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 1.92,
|
| 1144 |
+
"learning_rate": 4.3983958786933116e-05,
|
| 1145 |
+
"loss": 0.0296,
|
| 1146 |
+
"step": 7150
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 1.94,
|
| 1150 |
+
"learning_rate": 4.374461475696044e-05,
|
| 1151 |
+
"loss": 0.0363,
|
| 1152 |
+
"step": 7200
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 1.94,
|
| 1156 |
+
"eval_loss": 0.5993568897247314,
|
| 1157 |
+
"eval_runtime": 305.6394,
|
| 1158 |
+
"eval_samples_per_second": 76.976,
|
| 1159 |
+
"eval_steps_per_second": 9.622,
|
| 1160 |
+
"step": 7200
|
| 1161 |
+
}
|
| 1162 |
+
],
|
| 1163 |
+
"max_steps": 18575,
|
| 1164 |
+
"num_train_epochs": 5,
|
| 1165 |
+
"total_flos": 0.0,
|
| 1166 |
+
"trial_name": null,
|
| 1167 |
+
"trial_params": null
|
| 1168 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:de531913a431447efee6c7a4c834136608813a4d4813d16f2ffb9a26d0351a6b
|
| 3 |
+
size 3771
|