wu981526092
commited on
Commit
•
01c54c3
1
Parent(s):
7866a75
Upload 11 files
Browse files- model.safetensors +1 -1
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +15 -0
- spiece.model +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +57 -0
- trainer_state.json +1459 -0
- training_args.bin +3 -0
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 46743912
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5e96e57fb4c2ffe175277ebb65b5cd16a44a7114097c0e2e688b55fef924b13a
|
3 |
size 46743912
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0a504f4b6efd623f8cfa8d97bf7ceef00d87744ec09b0ee121c51cbf548033cd
|
3 |
+
size 93502808
|
rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b5109e08295ec457bd444c73df2b91ea091459c68e440d157572b04e7050331a
|
3 |
+
size 13553
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7477571713a4a349a0ecb64fd21f32fe71fd5559a2cbd3c32e4c358dca8bc7ce
|
3 |
+
size 627
|
special_tokens_map.json
ADDED
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "[CLS]",
|
3 |
+
"cls_token": "[CLS]",
|
4 |
+
"eos_token": "[SEP]",
|
5 |
+
"mask_token": {
|
6 |
+
"content": "[MASK]",
|
7 |
+
"lstrip": true,
|
8 |
+
"normalized": false,
|
9 |
+
"rstrip": false,
|
10 |
+
"single_word": false
|
11 |
+
},
|
12 |
+
"pad_token": "<pad>",
|
13 |
+
"sep_token": "[SEP]",
|
14 |
+
"unk_token": "<unk>"
|
15 |
+
}
|
spiece.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fefb02b667a6c5c2fe27602d28e5fb3428f66ab89c7d6f388e7c8d44a02d0336
|
3 |
+
size 760289
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "<pad>",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"1": {
|
12 |
+
"content": "<unk>",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"2": {
|
20 |
+
"content": "[CLS]",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
},
|
27 |
+
"3": {
|
28 |
+
"content": "[SEP]",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
},
|
35 |
+
"4": {
|
36 |
+
"content": "[MASK]",
|
37 |
+
"lstrip": true,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
}
|
43 |
+
},
|
44 |
+
"bos_token": "[CLS]",
|
45 |
+
"clean_up_tokenization_spaces": true,
|
46 |
+
"cls_token": "[CLS]",
|
47 |
+
"do_lower_case": true,
|
48 |
+
"eos_token": "[SEP]",
|
49 |
+
"keep_accents": false,
|
50 |
+
"mask_token": "[MASK]",
|
51 |
+
"model_max_length": 512,
|
52 |
+
"pad_token": "<pad>",
|
53 |
+
"remove_space": true,
|
54 |
+
"sep_token": "[SEP]",
|
55 |
+
"tokenizer_class": "AlbertTokenizer",
|
56 |
+
"unk_token": "<unk>"
|
57 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1459 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.29854172468185425,
|
3 |
+
"best_model_checkpoint": "./results/checkpoint-2026",
|
4 |
+
"epoch": 2.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 2026,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.01,
|
13 |
+
"grad_norm": 15.60984992980957,
|
14 |
+
"learning_rate": 4.990128331688055e-05,
|
15 |
+
"loss": 0.5627,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.02,
|
20 |
+
"grad_norm": 1.2870399951934814,
|
21 |
+
"learning_rate": 4.9802566633761114e-05,
|
22 |
+
"loss": 0.2714,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.03,
|
27 |
+
"grad_norm": 123.16035461425781,
|
28 |
+
"learning_rate": 4.970384995064166e-05,
|
29 |
+
"loss": 0.5078,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.04,
|
34 |
+
"grad_norm": 11.994287490844727,
|
35 |
+
"learning_rate": 4.960513326752221e-05,
|
36 |
+
"loss": 0.4197,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.05,
|
41 |
+
"grad_norm": 3.41953182220459,
|
42 |
+
"learning_rate": 4.950641658440277e-05,
|
43 |
+
"loss": 0.441,
|
44 |
+
"step": 50
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.06,
|
48 |
+
"grad_norm": 26.95296287536621,
|
49 |
+
"learning_rate": 4.940769990128332e-05,
|
50 |
+
"loss": 0.6958,
|
51 |
+
"step": 60
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.07,
|
55 |
+
"grad_norm": 28.05646324157715,
|
56 |
+
"learning_rate": 4.930898321816387e-05,
|
57 |
+
"loss": 0.3109,
|
58 |
+
"step": 70
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.08,
|
62 |
+
"grad_norm": 28.755550384521484,
|
63 |
+
"learning_rate": 4.921026653504443e-05,
|
64 |
+
"loss": 0.5321,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.09,
|
69 |
+
"grad_norm": 46.09171676635742,
|
70 |
+
"learning_rate": 4.9111549851924976e-05,
|
71 |
+
"loss": 0.4216,
|
72 |
+
"step": 90
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.1,
|
76 |
+
"grad_norm": 3.234527587890625,
|
77 |
+
"learning_rate": 4.901283316880553e-05,
|
78 |
+
"loss": 0.3427,
|
79 |
+
"step": 100
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.11,
|
83 |
+
"grad_norm": 36.78240966796875,
|
84 |
+
"learning_rate": 4.891411648568609e-05,
|
85 |
+
"loss": 0.5259,
|
86 |
+
"step": 110
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.12,
|
90 |
+
"grad_norm": 2.4952964782714844,
|
91 |
+
"learning_rate": 4.8815399802566636e-05,
|
92 |
+
"loss": 0.525,
|
93 |
+
"step": 120
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.13,
|
97 |
+
"grad_norm": 6.107447147369385,
|
98 |
+
"learning_rate": 4.8716683119447184e-05,
|
99 |
+
"loss": 0.5349,
|
100 |
+
"step": 130
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.14,
|
104 |
+
"grad_norm": 17.599472045898438,
|
105 |
+
"learning_rate": 4.861796643632775e-05,
|
106 |
+
"loss": 0.3194,
|
107 |
+
"step": 140
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.15,
|
111 |
+
"grad_norm": 67.53023529052734,
|
112 |
+
"learning_rate": 4.8519249753208296e-05,
|
113 |
+
"loss": 0.4738,
|
114 |
+
"step": 150
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.16,
|
118 |
+
"grad_norm": 61.95085525512695,
|
119 |
+
"learning_rate": 4.8420533070088844e-05,
|
120 |
+
"loss": 0.4151,
|
121 |
+
"step": 160
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.17,
|
125 |
+
"grad_norm": 377.9793701171875,
|
126 |
+
"learning_rate": 4.83218163869694e-05,
|
127 |
+
"loss": 0.3219,
|
128 |
+
"step": 170
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.18,
|
132 |
+
"grad_norm": 9.71474838256836,
|
133 |
+
"learning_rate": 4.8223099703849955e-05,
|
134 |
+
"loss": 0.2931,
|
135 |
+
"step": 180
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.19,
|
139 |
+
"grad_norm": 54.442691802978516,
|
140 |
+
"learning_rate": 4.8124383020730504e-05,
|
141 |
+
"loss": 0.3802,
|
142 |
+
"step": 190
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.2,
|
146 |
+
"grad_norm": 111.41837310791016,
|
147 |
+
"learning_rate": 4.802566633761106e-05,
|
148 |
+
"loss": 0.4909,
|
149 |
+
"step": 200
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.21,
|
153 |
+
"grad_norm": 28.207542419433594,
|
154 |
+
"learning_rate": 4.792694965449161e-05,
|
155 |
+
"loss": 0.392,
|
156 |
+
"step": 210
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.22,
|
160 |
+
"grad_norm": 65.766357421875,
|
161 |
+
"learning_rate": 4.7828232971372164e-05,
|
162 |
+
"loss": 0.3002,
|
163 |
+
"step": 220
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.23,
|
167 |
+
"grad_norm": 127.14469909667969,
|
168 |
+
"learning_rate": 4.772951628825272e-05,
|
169 |
+
"loss": 0.3654,
|
170 |
+
"step": 230
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.24,
|
174 |
+
"grad_norm": 0.19254250824451447,
|
175 |
+
"learning_rate": 4.763079960513327e-05,
|
176 |
+
"loss": 0.2953,
|
177 |
+
"step": 240
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.25,
|
181 |
+
"grad_norm": 30.106840133666992,
|
182 |
+
"learning_rate": 4.753208292201382e-05,
|
183 |
+
"loss": 0.4019,
|
184 |
+
"step": 250
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.26,
|
188 |
+
"grad_norm": 4.884279727935791,
|
189 |
+
"learning_rate": 4.743336623889438e-05,
|
190 |
+
"loss": 0.4152,
|
191 |
+
"step": 260
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.27,
|
195 |
+
"grad_norm": 71.0513916015625,
|
196 |
+
"learning_rate": 4.733464955577493e-05,
|
197 |
+
"loss": 0.6281,
|
198 |
+
"step": 270
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.28,
|
202 |
+
"grad_norm": 1.381753921508789,
|
203 |
+
"learning_rate": 4.723593287265548e-05,
|
204 |
+
"loss": 0.3224,
|
205 |
+
"step": 280
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.29,
|
209 |
+
"grad_norm": 1.0361205339431763,
|
210 |
+
"learning_rate": 4.713721618953603e-05,
|
211 |
+
"loss": 0.6375,
|
212 |
+
"step": 290
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.3,
|
216 |
+
"grad_norm": 1.2360197305679321,
|
217 |
+
"learning_rate": 4.703849950641659e-05,
|
218 |
+
"loss": 0.419,
|
219 |
+
"step": 300
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.31,
|
223 |
+
"grad_norm": 188.26495361328125,
|
224 |
+
"learning_rate": 4.693978282329714e-05,
|
225 |
+
"loss": 0.5303,
|
226 |
+
"step": 310
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.32,
|
230 |
+
"grad_norm": 0.14256739616394043,
|
231 |
+
"learning_rate": 4.684106614017769e-05,
|
232 |
+
"loss": 0.2615,
|
233 |
+
"step": 320
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.33,
|
237 |
+
"grad_norm": 63.93450927734375,
|
238 |
+
"learning_rate": 4.674234945705824e-05,
|
239 |
+
"loss": 0.4678,
|
240 |
+
"step": 330
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.34,
|
244 |
+
"grad_norm": 31.07522201538086,
|
245 |
+
"learning_rate": 4.66436327739388e-05,
|
246 |
+
"loss": 0.607,
|
247 |
+
"step": 340
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.35,
|
251 |
+
"grad_norm": 12.982345581054688,
|
252 |
+
"learning_rate": 4.654491609081935e-05,
|
253 |
+
"loss": 0.189,
|
254 |
+
"step": 350
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.36,
|
258 |
+
"grad_norm": 14.37088394165039,
|
259 |
+
"learning_rate": 4.64461994076999e-05,
|
260 |
+
"loss": 0.4341,
|
261 |
+
"step": 360
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.37,
|
265 |
+
"grad_norm": 8.185881614685059,
|
266 |
+
"learning_rate": 4.634748272458046e-05,
|
267 |
+
"loss": 0.5132,
|
268 |
+
"step": 370
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.38,
|
272 |
+
"grad_norm": 1.082980751991272,
|
273 |
+
"learning_rate": 4.624876604146101e-05,
|
274 |
+
"loss": 0.3491,
|
275 |
+
"step": 380
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.38,
|
279 |
+
"grad_norm": 12.5576753616333,
|
280 |
+
"learning_rate": 4.615004935834156e-05,
|
281 |
+
"loss": 0.7587,
|
282 |
+
"step": 390
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.39,
|
286 |
+
"grad_norm": 5.933102130889893,
|
287 |
+
"learning_rate": 4.605133267522212e-05,
|
288 |
+
"loss": 0.579,
|
289 |
+
"step": 400
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.4,
|
293 |
+
"grad_norm": 1.3454967737197876,
|
294 |
+
"learning_rate": 4.5952615992102666e-05,
|
295 |
+
"loss": 0.1932,
|
296 |
+
"step": 410
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.41,
|
300 |
+
"grad_norm": 17.171228408813477,
|
301 |
+
"learning_rate": 4.585389930898322e-05,
|
302 |
+
"loss": 0.3217,
|
303 |
+
"step": 420
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.42,
|
307 |
+
"grad_norm": 3.5590412616729736,
|
308 |
+
"learning_rate": 4.575518262586377e-05,
|
309 |
+
"loss": 0.2279,
|
310 |
+
"step": 430
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.43,
|
314 |
+
"grad_norm": 17.661069869995117,
|
315 |
+
"learning_rate": 4.5656465942744326e-05,
|
316 |
+
"loss": 0.2676,
|
317 |
+
"step": 440
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.44,
|
321 |
+
"grad_norm": 48.93571853637695,
|
322 |
+
"learning_rate": 4.5557749259624875e-05,
|
323 |
+
"loss": 0.5449,
|
324 |
+
"step": 450
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.45,
|
328 |
+
"grad_norm": 12.7286376953125,
|
329 |
+
"learning_rate": 4.545903257650543e-05,
|
330 |
+
"loss": 0.5127,
|
331 |
+
"step": 460
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.46,
|
335 |
+
"grad_norm": 51.88860321044922,
|
336 |
+
"learning_rate": 4.5360315893385986e-05,
|
337 |
+
"loss": 0.4794,
|
338 |
+
"step": 470
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.47,
|
342 |
+
"grad_norm": 18.063552856445312,
|
343 |
+
"learning_rate": 4.5261599210266535e-05,
|
344 |
+
"loss": 0.3728,
|
345 |
+
"step": 480
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.48,
|
349 |
+
"grad_norm": 2.861877918243408,
|
350 |
+
"learning_rate": 4.516288252714709e-05,
|
351 |
+
"loss": 0.3038,
|
352 |
+
"step": 490
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.49,
|
356 |
+
"grad_norm": 6.476074695587158,
|
357 |
+
"learning_rate": 4.5064165844027646e-05,
|
358 |
+
"loss": 0.2592,
|
359 |
+
"step": 500
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.5,
|
363 |
+
"grad_norm": 62.48997497558594,
|
364 |
+
"learning_rate": 4.4965449160908195e-05,
|
365 |
+
"loss": 0.4779,
|
366 |
+
"step": 510
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.51,
|
370 |
+
"grad_norm": 0.5959272384643555,
|
371 |
+
"learning_rate": 4.486673247778875e-05,
|
372 |
+
"loss": 0.3865,
|
373 |
+
"step": 520
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.52,
|
377 |
+
"grad_norm": 32.949684143066406,
|
378 |
+
"learning_rate": 4.47680157946693e-05,
|
379 |
+
"loss": 0.5077,
|
380 |
+
"step": 530
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.53,
|
384 |
+
"grad_norm": 8.09738826751709,
|
385 |
+
"learning_rate": 4.4669299111549855e-05,
|
386 |
+
"loss": 0.3352,
|
387 |
+
"step": 540
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.54,
|
391 |
+
"grad_norm": 23.277297973632812,
|
392 |
+
"learning_rate": 4.457058242843041e-05,
|
393 |
+
"loss": 0.5204,
|
394 |
+
"step": 550
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.55,
|
398 |
+
"grad_norm": 89.32869720458984,
|
399 |
+
"learning_rate": 4.447186574531096e-05,
|
400 |
+
"loss": 0.3888,
|
401 |
+
"step": 560
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.56,
|
405 |
+
"grad_norm": 2.6795363426208496,
|
406 |
+
"learning_rate": 4.437314906219151e-05,
|
407 |
+
"loss": 0.5252,
|
408 |
+
"step": 570
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.57,
|
412 |
+
"grad_norm": 37.583744049072266,
|
413 |
+
"learning_rate": 4.427443237907207e-05,
|
414 |
+
"loss": 0.3881,
|
415 |
+
"step": 580
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.58,
|
419 |
+
"grad_norm": 1.256844162940979,
|
420 |
+
"learning_rate": 4.417571569595262e-05,
|
421 |
+
"loss": 0.1872,
|
422 |
+
"step": 590
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.59,
|
426 |
+
"grad_norm": 4.5737786293029785,
|
427 |
+
"learning_rate": 4.407699901283317e-05,
|
428 |
+
"loss": 0.2536,
|
429 |
+
"step": 600
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.6,
|
433 |
+
"grad_norm": 45.64347839355469,
|
434 |
+
"learning_rate": 4.3978282329713724e-05,
|
435 |
+
"loss": 0.3777,
|
436 |
+
"step": 610
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.61,
|
440 |
+
"grad_norm": 0.4227633774280548,
|
441 |
+
"learning_rate": 4.387956564659428e-05,
|
442 |
+
"loss": 0.2028,
|
443 |
+
"step": 620
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.62,
|
447 |
+
"grad_norm": 4.602664947509766,
|
448 |
+
"learning_rate": 4.378084896347483e-05,
|
449 |
+
"loss": 0.5563,
|
450 |
+
"step": 630
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.63,
|
454 |
+
"grad_norm": 0.7803702354431152,
|
455 |
+
"learning_rate": 4.3682132280355384e-05,
|
456 |
+
"loss": 0.3636,
|
457 |
+
"step": 640
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.64,
|
461 |
+
"grad_norm": 70.02734375,
|
462 |
+
"learning_rate": 4.358341559723593e-05,
|
463 |
+
"loss": 0.4558,
|
464 |
+
"step": 650
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.65,
|
468 |
+
"grad_norm": 39.45964050292969,
|
469 |
+
"learning_rate": 4.348469891411649e-05,
|
470 |
+
"loss": 0.4592,
|
471 |
+
"step": 660
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.66,
|
475 |
+
"grad_norm": 22.5675106048584,
|
476 |
+
"learning_rate": 4.3385982230997044e-05,
|
477 |
+
"loss": 0.3082,
|
478 |
+
"step": 670
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 0.67,
|
482 |
+
"grad_norm": 4.789850234985352,
|
483 |
+
"learning_rate": 4.328726554787759e-05,
|
484 |
+
"loss": 0.2404,
|
485 |
+
"step": 680
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.68,
|
489 |
+
"grad_norm": 4.671356678009033,
|
490 |
+
"learning_rate": 4.318854886475814e-05,
|
491 |
+
"loss": 0.2864,
|
492 |
+
"step": 690
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.69,
|
496 |
+
"grad_norm": 1.803113341331482,
|
497 |
+
"learning_rate": 4.3089832181638704e-05,
|
498 |
+
"loss": 0.2627,
|
499 |
+
"step": 700
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.7,
|
503 |
+
"grad_norm": 0.38143932819366455,
|
504 |
+
"learning_rate": 4.299111549851925e-05,
|
505 |
+
"loss": 0.1678,
|
506 |
+
"step": 710
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.71,
|
510 |
+
"grad_norm": 0.396694540977478,
|
511 |
+
"learning_rate": 4.28923988153998e-05,
|
512 |
+
"loss": 0.181,
|
513 |
+
"step": 720
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.72,
|
517 |
+
"grad_norm": 26.724634170532227,
|
518 |
+
"learning_rate": 4.279368213228036e-05,
|
519 |
+
"loss": 0.5595,
|
520 |
+
"step": 730
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.73,
|
524 |
+
"grad_norm": 179.3428497314453,
|
525 |
+
"learning_rate": 4.269496544916091e-05,
|
526 |
+
"loss": 0.3613,
|
527 |
+
"step": 740
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.74,
|
531 |
+
"grad_norm": 4.721936225891113,
|
532 |
+
"learning_rate": 4.259624876604146e-05,
|
533 |
+
"loss": 0.4182,
|
534 |
+
"step": 750
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 0.75,
|
538 |
+
"grad_norm": 1.8950241804122925,
|
539 |
+
"learning_rate": 4.249753208292202e-05,
|
540 |
+
"loss": 0.3623,
|
541 |
+
"step": 760
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.76,
|
545 |
+
"grad_norm": 5.388864994049072,
|
546 |
+
"learning_rate": 4.2398815399802566e-05,
|
547 |
+
"loss": 0.4246,
|
548 |
+
"step": 770
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.77,
|
552 |
+
"grad_norm": 0.41123124957084656,
|
553 |
+
"learning_rate": 4.230009871668312e-05,
|
554 |
+
"loss": 0.2425,
|
555 |
+
"step": 780
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.78,
|
559 |
+
"grad_norm": 0.3556106388568878,
|
560 |
+
"learning_rate": 4.220138203356368e-05,
|
561 |
+
"loss": 0.3751,
|
562 |
+
"step": 790
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.79,
|
566 |
+
"grad_norm": 0.899945080280304,
|
567 |
+
"learning_rate": 4.2102665350444226e-05,
|
568 |
+
"loss": 0.3994,
|
569 |
+
"step": 800
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.8,
|
573 |
+
"grad_norm": 4.583869934082031,
|
574 |
+
"learning_rate": 4.2003948667324774e-05,
|
575 |
+
"loss": 0.3681,
|
576 |
+
"step": 810
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.81,
|
580 |
+
"grad_norm": 0.3905455768108368,
|
581 |
+
"learning_rate": 4.190523198420534e-05,
|
582 |
+
"loss": 0.1491,
|
583 |
+
"step": 820
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.82,
|
587 |
+
"grad_norm": 36.8359260559082,
|
588 |
+
"learning_rate": 4.1806515301085886e-05,
|
589 |
+
"loss": 0.2609,
|
590 |
+
"step": 830
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.83,
|
594 |
+
"grad_norm": 34.53616714477539,
|
595 |
+
"learning_rate": 4.1707798617966434e-05,
|
596 |
+
"loss": 0.5495,
|
597 |
+
"step": 840
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.84,
|
601 |
+
"grad_norm": 14.104715347290039,
|
602 |
+
"learning_rate": 4.160908193484699e-05,
|
603 |
+
"loss": 0.33,
|
604 |
+
"step": 850
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.85,
|
608 |
+
"grad_norm": 30.295068740844727,
|
609 |
+
"learning_rate": 4.1510365251727546e-05,
|
610 |
+
"loss": 1.0008,
|
611 |
+
"step": 860
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.86,
|
615 |
+
"grad_norm": 93.3653793334961,
|
616 |
+
"learning_rate": 4.1411648568608094e-05,
|
617 |
+
"loss": 1.0401,
|
618 |
+
"step": 870
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 0.87,
|
622 |
+
"grad_norm": 114.31365966796875,
|
623 |
+
"learning_rate": 4.131293188548865e-05,
|
624 |
+
"loss": 0.4156,
|
625 |
+
"step": 880
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.88,
|
629 |
+
"grad_norm": 134.54774475097656,
|
630 |
+
"learning_rate": 4.12142152023692e-05,
|
631 |
+
"loss": 0.5463,
|
632 |
+
"step": 890
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.89,
|
636 |
+
"grad_norm": 3.021076202392578,
|
637 |
+
"learning_rate": 4.1115498519249754e-05,
|
638 |
+
"loss": 0.2947,
|
639 |
+
"step": 900
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.9,
|
643 |
+
"grad_norm": 9.884215354919434,
|
644 |
+
"learning_rate": 4.101678183613031e-05,
|
645 |
+
"loss": 0.3674,
|
646 |
+
"step": 910
|
647 |
+
},
|
648 |
+
{
|
649 |
+
"epoch": 0.91,
|
650 |
+
"grad_norm": 167.9898223876953,
|
651 |
+
"learning_rate": 4.091806515301086e-05,
|
652 |
+
"loss": 0.4516,
|
653 |
+
"step": 920
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.92,
|
657 |
+
"grad_norm": 34.41691207885742,
|
658 |
+
"learning_rate": 4.0819348469891414e-05,
|
659 |
+
"loss": 0.504,
|
660 |
+
"step": 930
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 0.93,
|
664 |
+
"grad_norm": 10.135024070739746,
|
665 |
+
"learning_rate": 4.072063178677197e-05,
|
666 |
+
"loss": 0.2834,
|
667 |
+
"step": 940
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 0.94,
|
671 |
+
"grad_norm": 1.0688509941101074,
|
672 |
+
"learning_rate": 4.062191510365252e-05,
|
673 |
+
"loss": 0.3188,
|
674 |
+
"step": 950
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 0.95,
|
678 |
+
"grad_norm": 5.052711009979248,
|
679 |
+
"learning_rate": 4.052319842053307e-05,
|
680 |
+
"loss": 0.3693,
|
681 |
+
"step": 960
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.96,
|
685 |
+
"grad_norm": 0.37648436427116394,
|
686 |
+
"learning_rate": 4.042448173741363e-05,
|
687 |
+
"loss": 0.1054,
|
688 |
+
"step": 970
|
689 |
+
},
|
690 |
+
{
|
691 |
+
"epoch": 0.97,
|
692 |
+
"grad_norm": 18.3348445892334,
|
693 |
+
"learning_rate": 4.032576505429418e-05,
|
694 |
+
"loss": 0.3397,
|
695 |
+
"step": 980
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 0.98,
|
699 |
+
"grad_norm": 10.808074951171875,
|
700 |
+
"learning_rate": 4.022704837117473e-05,
|
701 |
+
"loss": 0.3628,
|
702 |
+
"step": 990
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 0.99,
|
706 |
+
"grad_norm": 141.88064575195312,
|
707 |
+
"learning_rate": 4.012833168805528e-05,
|
708 |
+
"loss": 0.9269,
|
709 |
+
"step": 1000
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 1.0,
|
713 |
+
"grad_norm": 0.555182695388794,
|
714 |
+
"learning_rate": 4.002961500493584e-05,
|
715 |
+
"loss": 0.1197,
|
716 |
+
"step": 1010
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 1.0,
|
720 |
+
"eval_balanced accuracy": 0.917760474601409,
|
721 |
+
"eval_f1": 0.9176981176842771,
|
722 |
+
"eval_loss": 0.40740078687667847,
|
723 |
+
"eval_precision": 0.9176448492816227,
|
724 |
+
"eval_recall": 0.917760474601409,
|
725 |
+
"eval_runtime": 5.5647,
|
726 |
+
"eval_samples_per_second": 161.733,
|
727 |
+
"eval_steps_per_second": 10.243,
|
728 |
+
"step": 1013
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 1.01,
|
732 |
+
"grad_norm": 4.549361228942871,
|
733 |
+
"learning_rate": 3.993089832181639e-05,
|
734 |
+
"loss": 0.5231,
|
735 |
+
"step": 1020
|
736 |
+
},
|
737 |
+
{
|
738 |
+
"epoch": 1.02,
|
739 |
+
"grad_norm": 5.699501991271973,
|
740 |
+
"learning_rate": 3.983218163869694e-05,
|
741 |
+
"loss": 0.4139,
|
742 |
+
"step": 1030
|
743 |
+
},
|
744 |
+
{
|
745 |
+
"epoch": 1.03,
|
746 |
+
"grad_norm": 2.1153147220611572,
|
747 |
+
"learning_rate": 3.973346495557749e-05,
|
748 |
+
"loss": 0.2718,
|
749 |
+
"step": 1040
|
750 |
+
},
|
751 |
+
{
|
752 |
+
"epoch": 1.04,
|
753 |
+
"grad_norm": 5.258866310119629,
|
754 |
+
"learning_rate": 3.963474827245805e-05,
|
755 |
+
"loss": 0.3115,
|
756 |
+
"step": 1050
|
757 |
+
},
|
758 |
+
{
|
759 |
+
"epoch": 1.05,
|
760 |
+
"grad_norm": 13.351494789123535,
|
761 |
+
"learning_rate": 3.95360315893386e-05,
|
762 |
+
"loss": 0.3992,
|
763 |
+
"step": 1060
|
764 |
+
},
|
765 |
+
{
|
766 |
+
"epoch": 1.06,
|
767 |
+
"grad_norm": 9.7189359664917,
|
768 |
+
"learning_rate": 3.943731490621915e-05,
|
769 |
+
"loss": 0.1346,
|
770 |
+
"step": 1070
|
771 |
+
},
|
772 |
+
{
|
773 |
+
"epoch": 1.07,
|
774 |
+
"grad_norm": 5.006288051605225,
|
775 |
+
"learning_rate": 3.933859822309971e-05,
|
776 |
+
"loss": 0.3118,
|
777 |
+
"step": 1080
|
778 |
+
},
|
779 |
+
{
|
780 |
+
"epoch": 1.08,
|
781 |
+
"grad_norm": 7.094489574432373,
|
782 |
+
"learning_rate": 3.923988153998026e-05,
|
783 |
+
"loss": 0.1807,
|
784 |
+
"step": 1090
|
785 |
+
},
|
786 |
+
{
|
787 |
+
"epoch": 1.09,
|
788 |
+
"grad_norm": 4.784492492675781,
|
789 |
+
"learning_rate": 3.914116485686081e-05,
|
790 |
+
"loss": 0.3839,
|
791 |
+
"step": 1100
|
792 |
+
},
|
793 |
+
{
|
794 |
+
"epoch": 1.1,
|
795 |
+
"grad_norm": 1.5643423795700073,
|
796 |
+
"learning_rate": 3.904244817374136e-05,
|
797 |
+
"loss": 0.1729,
|
798 |
+
"step": 1110
|
799 |
+
},
|
800 |
+
{
|
801 |
+
"epoch": 1.11,
|
802 |
+
"grad_norm": 33.595703125,
|
803 |
+
"learning_rate": 3.8943731490621916e-05,
|
804 |
+
"loss": 0.1749,
|
805 |
+
"step": 1120
|
806 |
+
},
|
807 |
+
{
|
808 |
+
"epoch": 1.12,
|
809 |
+
"grad_norm": 0.5887395143508911,
|
810 |
+
"learning_rate": 3.884501480750247e-05,
|
811 |
+
"loss": 0.2513,
|
812 |
+
"step": 1130
|
813 |
+
},
|
814 |
+
{
|
815 |
+
"epoch": 1.13,
|
816 |
+
"grad_norm": 22.53057289123535,
|
817 |
+
"learning_rate": 3.874629812438302e-05,
|
818 |
+
"loss": 0.2858,
|
819 |
+
"step": 1140
|
820 |
+
},
|
821 |
+
{
|
822 |
+
"epoch": 1.14,
|
823 |
+
"grad_norm": 52.66212463378906,
|
824 |
+
"learning_rate": 3.8647581441263576e-05,
|
825 |
+
"loss": 0.1328,
|
826 |
+
"step": 1150
|
827 |
+
},
|
828 |
+
{
|
829 |
+
"epoch": 1.15,
|
830 |
+
"grad_norm": 5.8826117515563965,
|
831 |
+
"learning_rate": 3.8548864758144125e-05,
|
832 |
+
"loss": 0.3296,
|
833 |
+
"step": 1160
|
834 |
+
},
|
835 |
+
{
|
836 |
+
"epoch": 1.15,
|
837 |
+
"grad_norm": 10.208854675292969,
|
838 |
+
"learning_rate": 3.845014807502468e-05,
|
839 |
+
"loss": 0.1743,
|
840 |
+
"step": 1170
|
841 |
+
},
|
842 |
+
{
|
843 |
+
"epoch": 1.16,
|
844 |
+
"grad_norm": 5.222922325134277,
|
845 |
+
"learning_rate": 3.8351431391905236e-05,
|
846 |
+
"loss": 0.2482,
|
847 |
+
"step": 1180
|
848 |
+
},
|
849 |
+
{
|
850 |
+
"epoch": 1.17,
|
851 |
+
"grad_norm": 0.3885471224784851,
|
852 |
+
"learning_rate": 3.8252714708785785e-05,
|
853 |
+
"loss": 0.3651,
|
854 |
+
"step": 1190
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 1.18,
|
858 |
+
"grad_norm": 68.36416625976562,
|
859 |
+
"learning_rate": 3.815399802566634e-05,
|
860 |
+
"loss": 0.5256,
|
861 |
+
"step": 1200
|
862 |
+
},
|
863 |
+
{
|
864 |
+
"epoch": 1.19,
|
865 |
+
"grad_norm": 103.91950988769531,
|
866 |
+
"learning_rate": 3.8055281342546896e-05,
|
867 |
+
"loss": 0.2199,
|
868 |
+
"step": 1210
|
869 |
+
},
|
870 |
+
{
|
871 |
+
"epoch": 1.2,
|
872 |
+
"grad_norm": 0.17333897948265076,
|
873 |
+
"learning_rate": 3.7956564659427445e-05,
|
874 |
+
"loss": 0.126,
|
875 |
+
"step": 1220
|
876 |
+
},
|
877 |
+
{
|
878 |
+
"epoch": 1.21,
|
879 |
+
"grad_norm": 41.487117767333984,
|
880 |
+
"learning_rate": 3.7857847976308e-05,
|
881 |
+
"loss": 0.2293,
|
882 |
+
"step": 1230
|
883 |
+
},
|
884 |
+
{
|
885 |
+
"epoch": 1.22,
|
886 |
+
"grad_norm": 0.1527445763349533,
|
887 |
+
"learning_rate": 3.775913129318855e-05,
|
888 |
+
"loss": 0.2754,
|
889 |
+
"step": 1240
|
890 |
+
},
|
891 |
+
{
|
892 |
+
"epoch": 1.23,
|
893 |
+
"grad_norm": 0.3720811605453491,
|
894 |
+
"learning_rate": 3.7660414610069105e-05,
|
895 |
+
"loss": 0.1904,
|
896 |
+
"step": 1250
|
897 |
+
},
|
898 |
+
{
|
899 |
+
"epoch": 1.24,
|
900 |
+
"grad_norm": 0.2801426947116852,
|
901 |
+
"learning_rate": 3.756169792694966e-05,
|
902 |
+
"loss": 0.2894,
|
903 |
+
"step": 1260
|
904 |
+
},
|
905 |
+
{
|
906 |
+
"epoch": 1.25,
|
907 |
+
"grad_norm": 0.912218451499939,
|
908 |
+
"learning_rate": 3.746298124383021e-05,
|
909 |
+
"loss": 0.4345,
|
910 |
+
"step": 1270
|
911 |
+
},
|
912 |
+
{
|
913 |
+
"epoch": 1.26,
|
914 |
+
"grad_norm": 0.25501587986946106,
|
915 |
+
"learning_rate": 3.736426456071076e-05,
|
916 |
+
"loss": 0.2249,
|
917 |
+
"step": 1280
|
918 |
+
},
|
919 |
+
{
|
920 |
+
"epoch": 1.27,
|
921 |
+
"grad_norm": 19.25888442993164,
|
922 |
+
"learning_rate": 3.7265547877591314e-05,
|
923 |
+
"loss": 0.4532,
|
924 |
+
"step": 1290
|
925 |
+
},
|
926 |
+
{
|
927 |
+
"epoch": 1.28,
|
928 |
+
"grad_norm": 7.447415351867676,
|
929 |
+
"learning_rate": 3.716683119447187e-05,
|
930 |
+
"loss": 0.419,
|
931 |
+
"step": 1300
|
932 |
+
},
|
933 |
+
{
|
934 |
+
"epoch": 1.29,
|
935 |
+
"grad_norm": 1.2623952627182007,
|
936 |
+
"learning_rate": 3.706811451135242e-05,
|
937 |
+
"loss": 0.3596,
|
938 |
+
"step": 1310
|
939 |
+
},
|
940 |
+
{
|
941 |
+
"epoch": 1.3,
|
942 |
+
"grad_norm": 49.27845001220703,
|
943 |
+
"learning_rate": 3.6969397828232974e-05,
|
944 |
+
"loss": 0.1807,
|
945 |
+
"step": 1320
|
946 |
+
},
|
947 |
+
{
|
948 |
+
"epoch": 1.31,
|
949 |
+
"grad_norm": 8.055280685424805,
|
950 |
+
"learning_rate": 3.687068114511353e-05,
|
951 |
+
"loss": 0.1877,
|
952 |
+
"step": 1330
|
953 |
+
},
|
954 |
+
{
|
955 |
+
"epoch": 1.32,
|
956 |
+
"grad_norm": 0.24801558256149292,
|
957 |
+
"learning_rate": 3.677196446199408e-05,
|
958 |
+
"loss": 0.1906,
|
959 |
+
"step": 1340
|
960 |
+
},
|
961 |
+
{
|
962 |
+
"epoch": 1.33,
|
963 |
+
"grad_norm": 0.37148603796958923,
|
964 |
+
"learning_rate": 3.6673247778874634e-05,
|
965 |
+
"loss": 0.6613,
|
966 |
+
"step": 1350
|
967 |
+
},
|
968 |
+
{
|
969 |
+
"epoch": 1.34,
|
970 |
+
"grad_norm": 2.0603933334350586,
|
971 |
+
"learning_rate": 3.657453109575518e-05,
|
972 |
+
"loss": 0.1717,
|
973 |
+
"step": 1360
|
974 |
+
},
|
975 |
+
{
|
976 |
+
"epoch": 1.35,
|
977 |
+
"grad_norm": 1.4730746746063232,
|
978 |
+
"learning_rate": 3.647581441263574e-05,
|
979 |
+
"loss": 0.3606,
|
980 |
+
"step": 1370
|
981 |
+
},
|
982 |
+
{
|
983 |
+
"epoch": 1.36,
|
984 |
+
"grad_norm": 11.129170417785645,
|
985 |
+
"learning_rate": 3.6377097729516294e-05,
|
986 |
+
"loss": 0.4668,
|
987 |
+
"step": 1380
|
988 |
+
},
|
989 |
+
{
|
990 |
+
"epoch": 1.37,
|
991 |
+
"grad_norm": 107.76866912841797,
|
992 |
+
"learning_rate": 3.627838104639684e-05,
|
993 |
+
"loss": 0.4248,
|
994 |
+
"step": 1390
|
995 |
+
},
|
996 |
+
{
|
997 |
+
"epoch": 1.38,
|
998 |
+
"grad_norm": 0.4574478566646576,
|
999 |
+
"learning_rate": 3.617966436327739e-05,
|
1000 |
+
"loss": 0.2463,
|
1001 |
+
"step": 1400
|
1002 |
+
},
|
1003 |
+
{
|
1004 |
+
"epoch": 1.39,
|
1005 |
+
"grad_norm": 9.523133277893066,
|
1006 |
+
"learning_rate": 3.6080947680157954e-05,
|
1007 |
+
"loss": 0.2986,
|
1008 |
+
"step": 1410
|
1009 |
+
},
|
1010 |
+
{
|
1011 |
+
"epoch": 1.4,
|
1012 |
+
"grad_norm": 724.2791137695312,
|
1013 |
+
"learning_rate": 3.59822309970385e-05,
|
1014 |
+
"loss": 0.1994,
|
1015 |
+
"step": 1420
|
1016 |
+
},
|
1017 |
+
{
|
1018 |
+
"epoch": 1.41,
|
1019 |
+
"grad_norm": 0.495822012424469,
|
1020 |
+
"learning_rate": 3.588351431391905e-05,
|
1021 |
+
"loss": 0.405,
|
1022 |
+
"step": 1430
|
1023 |
+
},
|
1024 |
+
{
|
1025 |
+
"epoch": 1.42,
|
1026 |
+
"grad_norm": 0.7077971696853638,
|
1027 |
+
"learning_rate": 3.578479763079961e-05,
|
1028 |
+
"loss": 0.3258,
|
1029 |
+
"step": 1440
|
1030 |
+
},
|
1031 |
+
{
|
1032 |
+
"epoch": 1.43,
|
1033 |
+
"grad_norm": 0.471545934677124,
|
1034 |
+
"learning_rate": 3.568608094768016e-05,
|
1035 |
+
"loss": 0.3381,
|
1036 |
+
"step": 1450
|
1037 |
+
},
|
1038 |
+
{
|
1039 |
+
"epoch": 1.44,
|
1040 |
+
"grad_norm": 160.64279174804688,
|
1041 |
+
"learning_rate": 3.558736426456071e-05,
|
1042 |
+
"loss": 0.4319,
|
1043 |
+
"step": 1460
|
1044 |
+
},
|
1045 |
+
{
|
1046 |
+
"epoch": 1.45,
|
1047 |
+
"grad_norm": 213.93475341796875,
|
1048 |
+
"learning_rate": 3.548864758144127e-05,
|
1049 |
+
"loss": 0.3506,
|
1050 |
+
"step": 1470
|
1051 |
+
},
|
1052 |
+
{
|
1053 |
+
"epoch": 1.46,
|
1054 |
+
"grad_norm": 0.5124903917312622,
|
1055 |
+
"learning_rate": 3.5389930898321816e-05,
|
1056 |
+
"loss": 0.261,
|
1057 |
+
"step": 1480
|
1058 |
+
},
|
1059 |
+
{
|
1060 |
+
"epoch": 1.47,
|
1061 |
+
"grad_norm": 0.2033979296684265,
|
1062 |
+
"learning_rate": 3.529121421520237e-05,
|
1063 |
+
"loss": 0.3329,
|
1064 |
+
"step": 1490
|
1065 |
+
},
|
1066 |
+
{
|
1067 |
+
"epoch": 1.48,
|
1068 |
+
"grad_norm": 0.14042626321315765,
|
1069 |
+
"learning_rate": 3.519249753208293e-05,
|
1070 |
+
"loss": 0.198,
|
1071 |
+
"step": 1500
|
1072 |
+
},
|
1073 |
+
{
|
1074 |
+
"epoch": 1.49,
|
1075 |
+
"grad_norm": 0.052474942058324814,
|
1076 |
+
"learning_rate": 3.5093780848963476e-05,
|
1077 |
+
"loss": 0.3291,
|
1078 |
+
"step": 1510
|
1079 |
+
},
|
1080 |
+
{
|
1081 |
+
"epoch": 1.5,
|
1082 |
+
"grad_norm": 0.7498096823692322,
|
1083 |
+
"learning_rate": 3.4995064165844024e-05,
|
1084 |
+
"loss": 0.5893,
|
1085 |
+
"step": 1520
|
1086 |
+
},
|
1087 |
+
{
|
1088 |
+
"epoch": 1.51,
|
1089 |
+
"grad_norm": 56.467071533203125,
|
1090 |
+
"learning_rate": 3.489634748272459e-05,
|
1091 |
+
"loss": 0.22,
|
1092 |
+
"step": 1530
|
1093 |
+
},
|
1094 |
+
{
|
1095 |
+
"epoch": 1.52,
|
1096 |
+
"grad_norm": 5.047154903411865,
|
1097 |
+
"learning_rate": 3.4797630799605136e-05,
|
1098 |
+
"loss": 0.3128,
|
1099 |
+
"step": 1540
|
1100 |
+
},
|
1101 |
+
{
|
1102 |
+
"epoch": 1.53,
|
1103 |
+
"grad_norm": 0.24173791706562042,
|
1104 |
+
"learning_rate": 3.4698914116485684e-05,
|
1105 |
+
"loss": 0.2632,
|
1106 |
+
"step": 1550
|
1107 |
+
},
|
1108 |
+
{
|
1109 |
+
"epoch": 1.54,
|
1110 |
+
"grad_norm": 0.23745213449001312,
|
1111 |
+
"learning_rate": 3.460019743336624e-05,
|
1112 |
+
"loss": 0.1316,
|
1113 |
+
"step": 1560
|
1114 |
+
},
|
1115 |
+
{
|
1116 |
+
"epoch": 1.55,
|
1117 |
+
"grad_norm": 0.3697431683540344,
|
1118 |
+
"learning_rate": 3.4501480750246796e-05,
|
1119 |
+
"loss": 0.3162,
|
1120 |
+
"step": 1570
|
1121 |
+
},
|
1122 |
+
{
|
1123 |
+
"epoch": 1.56,
|
1124 |
+
"grad_norm": 125.36990356445312,
|
1125 |
+
"learning_rate": 3.4402764067127344e-05,
|
1126 |
+
"loss": 0.7252,
|
1127 |
+
"step": 1580
|
1128 |
+
},
|
1129 |
+
{
|
1130 |
+
"epoch": 1.57,
|
1131 |
+
"grad_norm": 30.01531410217285,
|
1132 |
+
"learning_rate": 3.43040473840079e-05,
|
1133 |
+
"loss": 0.567,
|
1134 |
+
"step": 1590
|
1135 |
+
},
|
1136 |
+
{
|
1137 |
+
"epoch": 1.58,
|
1138 |
+
"grad_norm": 44.524818420410156,
|
1139 |
+
"learning_rate": 3.420533070088845e-05,
|
1140 |
+
"loss": 0.4531,
|
1141 |
+
"step": 1600
|
1142 |
+
},
|
1143 |
+
{
|
1144 |
+
"epoch": 1.59,
|
1145 |
+
"grad_norm": 133.4363555908203,
|
1146 |
+
"learning_rate": 3.4106614017769004e-05,
|
1147 |
+
"loss": 0.4438,
|
1148 |
+
"step": 1610
|
1149 |
+
},
|
1150 |
+
{
|
1151 |
+
"epoch": 1.6,
|
1152 |
+
"grad_norm": 1119.47509765625,
|
1153 |
+
"learning_rate": 3.400789733464956e-05,
|
1154 |
+
"loss": 0.3973,
|
1155 |
+
"step": 1620
|
1156 |
+
},
|
1157 |
+
{
|
1158 |
+
"epoch": 1.61,
|
1159 |
+
"grad_norm": 4.369329929351807,
|
1160 |
+
"learning_rate": 3.390918065153011e-05,
|
1161 |
+
"loss": 0.482,
|
1162 |
+
"step": 1630
|
1163 |
+
},
|
1164 |
+
{
|
1165 |
+
"epoch": 1.62,
|
1166 |
+
"grad_norm": 28.413909912109375,
|
1167 |
+
"learning_rate": 3.381046396841066e-05,
|
1168 |
+
"loss": 0.3454,
|
1169 |
+
"step": 1640
|
1170 |
+
},
|
1171 |
+
{
|
1172 |
+
"epoch": 1.63,
|
1173 |
+
"grad_norm": 76.58002471923828,
|
1174 |
+
"learning_rate": 3.371174728529122e-05,
|
1175 |
+
"loss": 0.2663,
|
1176 |
+
"step": 1650
|
1177 |
+
},
|
1178 |
+
{
|
1179 |
+
"epoch": 1.64,
|
1180 |
+
"grad_norm": 597.3102416992188,
|
1181 |
+
"learning_rate": 3.361303060217177e-05,
|
1182 |
+
"loss": 0.155,
|
1183 |
+
"step": 1660
|
1184 |
+
},
|
1185 |
+
{
|
1186 |
+
"epoch": 1.65,
|
1187 |
+
"grad_norm": 24.984447479248047,
|
1188 |
+
"learning_rate": 3.351431391905232e-05,
|
1189 |
+
"loss": 0.2535,
|
1190 |
+
"step": 1670
|
1191 |
+
},
|
1192 |
+
{
|
1193 |
+
"epoch": 1.66,
|
1194 |
+
"grad_norm": 30.53813934326172,
|
1195 |
+
"learning_rate": 3.341559723593287e-05,
|
1196 |
+
"loss": 0.315,
|
1197 |
+
"step": 1680
|
1198 |
+
},
|
1199 |
+
{
|
1200 |
+
"epoch": 1.67,
|
1201 |
+
"grad_norm": 1.5513701438903809,
|
1202 |
+
"learning_rate": 3.331688055281343e-05,
|
1203 |
+
"loss": 0.3617,
|
1204 |
+
"step": 1690
|
1205 |
+
},
|
1206 |
+
{
|
1207 |
+
"epoch": 1.68,
|
1208 |
+
"grad_norm": 3.676360845565796,
|
1209 |
+
"learning_rate": 3.321816386969398e-05,
|
1210 |
+
"loss": 0.6472,
|
1211 |
+
"step": 1700
|
1212 |
+
},
|
1213 |
+
{
|
1214 |
+
"epoch": 1.69,
|
1215 |
+
"grad_norm": 23.96689796447754,
|
1216 |
+
"learning_rate": 3.311944718657453e-05,
|
1217 |
+
"loss": 0.5382,
|
1218 |
+
"step": 1710
|
1219 |
+
},
|
1220 |
+
{
|
1221 |
+
"epoch": 1.7,
|
1222 |
+
"grad_norm": 18.116992950439453,
|
1223 |
+
"learning_rate": 3.302073050345508e-05,
|
1224 |
+
"loss": 0.345,
|
1225 |
+
"step": 1720
|
1226 |
+
},
|
1227 |
+
{
|
1228 |
+
"epoch": 1.71,
|
1229 |
+
"grad_norm": 4.786412239074707,
|
1230 |
+
"learning_rate": 3.292201382033564e-05,
|
1231 |
+
"loss": 0.3599,
|
1232 |
+
"step": 1730
|
1233 |
+
},
|
1234 |
+
{
|
1235 |
+
"epoch": 1.72,
|
1236 |
+
"grad_norm": 2.5227644443511963,
|
1237 |
+
"learning_rate": 3.282329713721619e-05,
|
1238 |
+
"loss": 0.4313,
|
1239 |
+
"step": 1740
|
1240 |
+
},
|
1241 |
+
{
|
1242 |
+
"epoch": 1.73,
|
1243 |
+
"grad_norm": 4.462274074554443,
|
1244 |
+
"learning_rate": 3.272458045409674e-05,
|
1245 |
+
"loss": 0.5479,
|
1246 |
+
"step": 1750
|
1247 |
+
},
|
1248 |
+
{
|
1249 |
+
"epoch": 1.74,
|
1250 |
+
"grad_norm": 49.19129180908203,
|
1251 |
+
"learning_rate": 3.26258637709773e-05,
|
1252 |
+
"loss": 0.5215,
|
1253 |
+
"step": 1760
|
1254 |
+
},
|
1255 |
+
{
|
1256 |
+
"epoch": 1.75,
|
1257 |
+
"grad_norm": 89.65460968017578,
|
1258 |
+
"learning_rate": 3.252714708785785e-05,
|
1259 |
+
"loss": 0.7555,
|
1260 |
+
"step": 1770
|
1261 |
+
},
|
1262 |
+
{
|
1263 |
+
"epoch": 1.76,
|
1264 |
+
"grad_norm": 0.9293081760406494,
|
1265 |
+
"learning_rate": 3.24284304047384e-05,
|
1266 |
+
"loss": 0.3071,
|
1267 |
+
"step": 1780
|
1268 |
+
},
|
1269 |
+
{
|
1270 |
+
"epoch": 1.77,
|
1271 |
+
"grad_norm": 11.949310302734375,
|
1272 |
+
"learning_rate": 3.232971372161895e-05,
|
1273 |
+
"loss": 0.2182,
|
1274 |
+
"step": 1790
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 1.78,
|
1278 |
+
"grad_norm": 15.446320533752441,
|
1279 |
+
"learning_rate": 3.2230997038499506e-05,
|
1280 |
+
"loss": 0.2696,
|
1281 |
+
"step": 1800
|
1282 |
+
},
|
1283 |
+
{
|
1284 |
+
"epoch": 1.79,
|
1285 |
+
"grad_norm": 5.7437567710876465,
|
1286 |
+
"learning_rate": 3.213228035538006e-05,
|
1287 |
+
"loss": 0.3771,
|
1288 |
+
"step": 1810
|
1289 |
+
},
|
1290 |
+
{
|
1291 |
+
"epoch": 1.8,
|
1292 |
+
"grad_norm": 209.53298950195312,
|
1293 |
+
"learning_rate": 3.203356367226061e-05,
|
1294 |
+
"loss": 0.3023,
|
1295 |
+
"step": 1820
|
1296 |
+
},
|
1297 |
+
{
|
1298 |
+
"epoch": 1.81,
|
1299 |
+
"grad_norm": 1.2472151517868042,
|
1300 |
+
"learning_rate": 3.1934846989141166e-05,
|
1301 |
+
"loss": 0.29,
|
1302 |
+
"step": 1830
|
1303 |
+
},
|
1304 |
+
{
|
1305 |
+
"epoch": 1.82,
|
1306 |
+
"grad_norm": 370.38800048828125,
|
1307 |
+
"learning_rate": 3.1836130306021715e-05,
|
1308 |
+
"loss": 0.3409,
|
1309 |
+
"step": 1840
|
1310 |
+
},
|
1311 |
+
{
|
1312 |
+
"epoch": 1.83,
|
1313 |
+
"grad_norm": 145.07717895507812,
|
1314 |
+
"learning_rate": 3.173741362290227e-05,
|
1315 |
+
"loss": 0.3839,
|
1316 |
+
"step": 1850
|
1317 |
+
},
|
1318 |
+
{
|
1319 |
+
"epoch": 1.84,
|
1320 |
+
"grad_norm": 48.441585540771484,
|
1321 |
+
"learning_rate": 3.1638696939782826e-05,
|
1322 |
+
"loss": 0.2765,
|
1323 |
+
"step": 1860
|
1324 |
+
},
|
1325 |
+
{
|
1326 |
+
"epoch": 1.85,
|
1327 |
+
"grad_norm": 8.114079475402832,
|
1328 |
+
"learning_rate": 3.1539980256663375e-05,
|
1329 |
+
"loss": 0.4797,
|
1330 |
+
"step": 1870
|
1331 |
+
},
|
1332 |
+
{
|
1333 |
+
"epoch": 1.86,
|
1334 |
+
"grad_norm": 2.0335161685943604,
|
1335 |
+
"learning_rate": 3.144126357354393e-05,
|
1336 |
+
"loss": 0.3283,
|
1337 |
+
"step": 1880
|
1338 |
+
},
|
1339 |
+
{
|
1340 |
+
"epoch": 1.87,
|
1341 |
+
"grad_norm": 571.5001831054688,
|
1342 |
+
"learning_rate": 3.1342546890424486e-05,
|
1343 |
+
"loss": 0.3749,
|
1344 |
+
"step": 1890
|
1345 |
+
},
|
1346 |
+
{
|
1347 |
+
"epoch": 1.88,
|
1348 |
+
"grad_norm": 26.4891414642334,
|
1349 |
+
"learning_rate": 3.1243830207305035e-05,
|
1350 |
+
"loss": 0.5855,
|
1351 |
+
"step": 1900
|
1352 |
+
},
|
1353 |
+
{
|
1354 |
+
"epoch": 1.89,
|
1355 |
+
"grad_norm": 373.5781555175781,
|
1356 |
+
"learning_rate": 3.114511352418559e-05,
|
1357 |
+
"loss": 0.5779,
|
1358 |
+
"step": 1910
|
1359 |
+
},
|
1360 |
+
{
|
1361 |
+
"epoch": 1.9,
|
1362 |
+
"grad_norm": 12.153056144714355,
|
1363 |
+
"learning_rate": 3.1046396841066146e-05,
|
1364 |
+
"loss": 0.5083,
|
1365 |
+
"step": 1920
|
1366 |
+
},
|
1367 |
+
{
|
1368 |
+
"epoch": 1.91,
|
1369 |
+
"grad_norm": 246.8365936279297,
|
1370 |
+
"learning_rate": 3.0947680157946695e-05,
|
1371 |
+
"loss": 0.2794,
|
1372 |
+
"step": 1930
|
1373 |
+
},
|
1374 |
+
{
|
1375 |
+
"epoch": 1.92,
|
1376 |
+
"grad_norm": 16.38204002380371,
|
1377 |
+
"learning_rate": 3.084896347482725e-05,
|
1378 |
+
"loss": 0.1689,
|
1379 |
+
"step": 1940
|
1380 |
+
},
|
1381 |
+
{
|
1382 |
+
"epoch": 1.92,
|
1383 |
+
"grad_norm": 86.90618896484375,
|
1384 |
+
"learning_rate": 3.07502467917078e-05,
|
1385 |
+
"loss": 0.124,
|
1386 |
+
"step": 1950
|
1387 |
+
},
|
1388 |
+
{
|
1389 |
+
"epoch": 1.93,
|
1390 |
+
"grad_norm": 13.767565727233887,
|
1391 |
+
"learning_rate": 3.0651530108588355e-05,
|
1392 |
+
"loss": 0.4286,
|
1393 |
+
"step": 1960
|
1394 |
+
},
|
1395 |
+
{
|
1396 |
+
"epoch": 1.94,
|
1397 |
+
"grad_norm": 25.554912567138672,
|
1398 |
+
"learning_rate": 3.0552813425468904e-05,
|
1399 |
+
"loss": 0.342,
|
1400 |
+
"step": 1970
|
1401 |
+
},
|
1402 |
+
{
|
1403 |
+
"epoch": 1.95,
|
1404 |
+
"grad_norm": 0.8774542212486267,
|
1405 |
+
"learning_rate": 3.045409674234946e-05,
|
1406 |
+
"loss": 0.2024,
|
1407 |
+
"step": 1980
|
1408 |
+
},
|
1409 |
+
{
|
1410 |
+
"epoch": 1.96,
|
1411 |
+
"grad_norm": 53.33576583862305,
|
1412 |
+
"learning_rate": 3.0355380059230008e-05,
|
1413 |
+
"loss": 0.3032,
|
1414 |
+
"step": 1990
|
1415 |
+
},
|
1416 |
+
{
|
1417 |
+
"epoch": 1.97,
|
1418 |
+
"grad_norm": 6.252757549285889,
|
1419 |
+
"learning_rate": 3.0256663376110567e-05,
|
1420 |
+
"loss": 0.3633,
|
1421 |
+
"step": 2000
|
1422 |
+
},
|
1423 |
+
{
|
1424 |
+
"epoch": 1.98,
|
1425 |
+
"grad_norm": 0.8560687899589539,
|
1426 |
+
"learning_rate": 3.015794669299112e-05,
|
1427 |
+
"loss": 0.2662,
|
1428 |
+
"step": 2010
|
1429 |
+
},
|
1430 |
+
{
|
1431 |
+
"epoch": 1.99,
|
1432 |
+
"grad_norm": 5.560061454772949,
|
1433 |
+
"learning_rate": 3.0059230009871668e-05,
|
1434 |
+
"loss": 0.3057,
|
1435 |
+
"step": 2020
|
1436 |
+
},
|
1437 |
+
{
|
1438 |
+
"epoch": 2.0,
|
1439 |
+
"eval_balanced accuracy": 0.9113459399332592,
|
1440 |
+
"eval_f1": 0.9119394500117044,
|
1441 |
+
"eval_loss": 0.29854172468185425,
|
1442 |
+
"eval_precision": 0.913535516192521,
|
1443 |
+
"eval_recall": 0.9113459399332592,
|
1444 |
+
"eval_runtime": 5.6113,
|
1445 |
+
"eval_samples_per_second": 160.392,
|
1446 |
+
"eval_steps_per_second": 10.158,
|
1447 |
+
"step": 2026
|
1448 |
+
}
|
1449 |
+
],
|
1450 |
+
"logging_steps": 10,
|
1451 |
+
"max_steps": 5065,
|
1452 |
+
"num_input_tokens_seen": 0,
|
1453 |
+
"num_train_epochs": 5,
|
1454 |
+
"save_steps": 500,
|
1455 |
+
"total_flos": 96787312128000.0,
|
1456 |
+
"train_batch_size": 8,
|
1457 |
+
"trial_name": null,
|
1458 |
+
"trial_params": null
|
1459 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ce4159defbc8b7143d7a714e34b4537555801075b0d7f6b279e94c452fb5607c
|
3 |
+
size 4411
|