Upload folder using huggingface_hub
Browse files- adapter_model.safetensors +1 -1
- optimizer.pt +1 -1
- rng_state.pth +1 -1
- scheduler.pt +1 -1
- tokenizer.json +13 -17
- tokenizer_config.json +32 -24
- trainer_state.json +829 -3
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 16794200
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:709ae92384d7aeae2a594b24f20bc72a9ede6fc892d384e3edb835e78c4cc441
|
3 |
size 16794200
|
optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 33630266
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c8412229afcfd9ba03f3adc09a87546ff7c8ef8b4093a3684805704a3aa73147
|
3 |
size 33630266
|
rng_state.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14244
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b3cfd0f149ee197292991a77959f8ad42f317393eec06f6d5290905f53ad8df3
|
3 |
size 14244
|
scheduler.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1064
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7a1760d654783fa21ab023433f66d210cdfce9d79e1ba7107ab45b42aa3b2970
|
3 |
size 1064
|
tokenizer.json
CHANGED
@@ -1,6 +1,11 @@
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
-
"truncation":
|
|
|
|
|
|
|
|
|
|
|
4 |
"padding": null,
|
5 |
"added_tokens": [
|
6 |
{
|
@@ -31,23 +36,13 @@
|
|
31 |
"special": true
|
32 |
}
|
33 |
],
|
34 |
-
"normalizer":
|
35 |
-
|
36 |
-
"
|
37 |
-
|
38 |
-
|
39 |
-
|
40 |
-
},
|
41 |
-
{
|
42 |
-
"type": "Replace",
|
43 |
-
"pattern": {
|
44 |
-
"String": " "
|
45 |
-
},
|
46 |
-
"content": "▁"
|
47 |
-
}
|
48 |
-
]
|
49 |
},
|
50 |
-
"pre_tokenizer": null,
|
51 |
"post_processor": {
|
52 |
"type": "TemplateProcessing",
|
53 |
"single": [
|
@@ -134,6 +129,7 @@
|
|
134 |
"end_of_word_suffix": null,
|
135 |
"fuse_unk": true,
|
136 |
"byte_fallback": true,
|
|
|
137 |
"vocab": {
|
138 |
"<unk>": 0,
|
139 |
"<s>": 1,
|
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
+
"truncation": {
|
4 |
+
"direction": "Right",
|
5 |
+
"max_length": 1024,
|
6 |
+
"strategy": "LongestFirst",
|
7 |
+
"stride": 0
|
8 |
+
},
|
9 |
"padding": null,
|
10 |
"added_tokens": [
|
11 |
{
|
|
|
36 |
"special": true
|
37 |
}
|
38 |
],
|
39 |
+
"normalizer": null,
|
40 |
+
"pre_tokenizer": {
|
41 |
+
"type": "Metaspace",
|
42 |
+
"replacement": "▁",
|
43 |
+
"prepend_scheme": "first",
|
44 |
+
"split": false
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
45 |
},
|
|
|
46 |
"post_processor": {
|
47 |
"type": "TemplateProcessing",
|
48 |
"single": [
|
|
|
129 |
"end_of_word_suffix": null,
|
130 |
"fuse_unk": true,
|
131 |
"byte_fallback": true,
|
132 |
+
"ignore_merges": false,
|
133 |
"vocab": {
|
134 |
"<unk>": 0,
|
135 |
"<s>": 1,
|
tokenizer_config.json
CHANGED
@@ -1,35 +1,43 @@
|
|
1 |
{
|
2 |
"add_bos_token": true,
|
3 |
"add_eos_token": false,
|
4 |
-
"
|
5 |
-
|
6 |
-
"
|
7 |
-
|
8 |
-
|
9 |
-
|
10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
},
|
|
|
12 |
"clean_up_tokenization_spaces": false,
|
13 |
-
"eos_token":
|
14 |
-
"__type": "AddedToken",
|
15 |
-
"content": "</s>",
|
16 |
-
"lstrip": false,
|
17 |
-
"normalized": false,
|
18 |
-
"rstrip": false,
|
19 |
-
"single_word": false
|
20 |
-
},
|
21 |
"legacy": false,
|
22 |
"model_max_length": 4096,
|
23 |
-
"pad_token":
|
24 |
"padding_side": "right",
|
25 |
"sp_model_kwargs": {},
|
|
|
26 |
"tokenizer_class": "LlamaTokenizer",
|
27 |
-
"unk_token":
|
28 |
-
|
29 |
-
"content": "<unk>",
|
30 |
-
"lstrip": false,
|
31 |
-
"normalized": false,
|
32 |
-
"rstrip": false,
|
33 |
-
"single_word": false
|
34 |
-
}
|
35 |
}
|
|
|
1 |
{
|
2 |
"add_bos_token": true,
|
3 |
"add_eos_token": false,
|
4 |
+
"add_prefix_space": true,
|
5 |
+
"added_tokens_decoder": {
|
6 |
+
"0": {
|
7 |
+
"content": "<unk>",
|
8 |
+
"lstrip": false,
|
9 |
+
"normalized": false,
|
10 |
+
"rstrip": false,
|
11 |
+
"single_word": false,
|
12 |
+
"special": true
|
13 |
+
},
|
14 |
+
"1": {
|
15 |
+
"content": "<s>",
|
16 |
+
"lstrip": false,
|
17 |
+
"normalized": false,
|
18 |
+
"rstrip": false,
|
19 |
+
"single_word": false,
|
20 |
+
"special": true
|
21 |
+
},
|
22 |
+
"2": {
|
23 |
+
"content": "</s>",
|
24 |
+
"lstrip": false,
|
25 |
+
"normalized": false,
|
26 |
+
"rstrip": false,
|
27 |
+
"single_word": false,
|
28 |
+
"special": true
|
29 |
+
}
|
30 |
},
|
31 |
+
"bos_token": "<s>",
|
32 |
"clean_up_tokenization_spaces": false,
|
33 |
+
"eos_token": "</s>",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
34 |
"legacy": false,
|
35 |
"model_max_length": 4096,
|
36 |
+
"pad_token": "</s>",
|
37 |
"padding_side": "right",
|
38 |
"sp_model_kwargs": {},
|
39 |
+
"spaces_between_special_tokens": false,
|
40 |
"tokenizer_class": "LlamaTokenizer",
|
41 |
+
"unk_token": "<unk>",
|
42 |
+
"use_default_system_prompt": false
|
|
|
|
|
|
|
|
|
|
|
|
|
43 |
}
|
trainer_state.json
CHANGED
@@ -1,9 +1,9 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch":
|
5 |
"eval_steps": 500,
|
6 |
-
"global_step":
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
@@ -826,6 +826,832 @@
|
|
826 |
"learning_rate": 0.00015038541704796003,
|
827 |
"loss": 0.0139,
|
828 |
"step": 1170
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
829 |
}
|
830 |
],
|
831 |
"logging_steps": 10,
|
@@ -845,7 +1671,7 @@
|
|
845 |
"attributes": {}
|
846 |
}
|
847 |
},
|
848 |
-
"total_flos":
|
849 |
"train_batch_size": 1,
|
850 |
"trial_name": null,
|
851 |
"trial_params": null
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 10.0,
|
5 |
"eval_steps": 500,
|
6 |
+
"global_step": 2350,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
|
|
826 |
"learning_rate": 0.00015038541704796003,
|
827 |
"loss": 0.0139,
|
828 |
"step": 1170
|
829 |
+
},
|
830 |
+
{
|
831 |
+
"epoch": 5.0212765957446805,
|
832 |
+
"grad_norm": 0.2920892834663391,
|
833 |
+
"learning_rate": 0.0001496135900856782,
|
834 |
+
"loss": 0.0083,
|
835 |
+
"step": 1180
|
836 |
+
},
|
837 |
+
{
|
838 |
+
"epoch": 5.0638297872340425,
|
839 |
+
"grad_norm": 0.32393306493759155,
|
840 |
+
"learning_rate": 0.0001488378223697851,
|
841 |
+
"loss": 0.0155,
|
842 |
+
"step": 1190
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 5.1063829787234045,
|
846 |
+
"grad_norm": 0.49029844999313354,
|
847 |
+
"learning_rate": 0.00014805817551866838,
|
848 |
+
"loss": 0.0109,
|
849 |
+
"step": 1200
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 5.148936170212766,
|
853 |
+
"grad_norm": 0.05497799441218376,
|
854 |
+
"learning_rate": 0.00014727471145883127,
|
855 |
+
"loss": 0.0095,
|
856 |
+
"step": 1210
|
857 |
+
},
|
858 |
+
{
|
859 |
+
"epoch": 5.191489361702128,
|
860 |
+
"grad_norm": 0.4540445804595947,
|
861 |
+
"learning_rate": 0.00014648749241997363,
|
862 |
+
"loss": 0.0106,
|
863 |
+
"step": 1220
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 5.23404255319149,
|
867 |
+
"grad_norm": 0.16598157584667206,
|
868 |
+
"learning_rate": 0.00014569658093004935,
|
869 |
+
"loss": 0.0072,
|
870 |
+
"step": 1230
|
871 |
+
},
|
872 |
+
{
|
873 |
+
"epoch": 5.276595744680851,
|
874 |
+
"grad_norm": 0.07160704582929611,
|
875 |
+
"learning_rate": 0.0001449020398102996,
|
876 |
+
"loss": 0.0108,
|
877 |
+
"step": 1240
|
878 |
+
},
|
879 |
+
{
|
880 |
+
"epoch": 5.319148936170213,
|
881 |
+
"grad_norm": 0.197789266705513,
|
882 |
+
"learning_rate": 0.00014410393217026318,
|
883 |
+
"loss": 0.0118,
|
884 |
+
"step": 1250
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 5.361702127659575,
|
888 |
+
"grad_norm": 0.07983817905187607,
|
889 |
+
"learning_rate": 0.00014330232140276366,
|
890 |
+
"loss": 0.0076,
|
891 |
+
"step": 1260
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 5.404255319148936,
|
895 |
+
"grad_norm": 0.0746329054236412,
|
896 |
+
"learning_rate": 0.00014249727117887425,
|
897 |
+
"loss": 0.0089,
|
898 |
+
"step": 1270
|
899 |
+
},
|
900 |
+
{
|
901 |
+
"epoch": 5.446808510638298,
|
902 |
+
"grad_norm": 0.09392493963241577,
|
903 |
+
"learning_rate": 0.00014168884544286053,
|
904 |
+
"loss": 0.0103,
|
905 |
+
"step": 1280
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 5.48936170212766,
|
909 |
+
"grad_norm": 0.18386998772621155,
|
910 |
+
"learning_rate": 0.0001408771084071012,
|
911 |
+
"loss": 0.0096,
|
912 |
+
"step": 1290
|
913 |
+
},
|
914 |
+
{
|
915 |
+
"epoch": 5.531914893617021,
|
916 |
+
"grad_norm": 0.07400283962488174,
|
917 |
+
"learning_rate": 0.00014006212454698797,
|
918 |
+
"loss": 0.0083,
|
919 |
+
"step": 1300
|
920 |
+
},
|
921 |
+
{
|
922 |
+
"epoch": 5.574468085106383,
|
923 |
+
"grad_norm": 0.06513144075870514,
|
924 |
+
"learning_rate": 0.00013924395859580432,
|
925 |
+
"loss": 0.0093,
|
926 |
+
"step": 1310
|
927 |
+
},
|
928 |
+
{
|
929 |
+
"epoch": 5.617021276595745,
|
930 |
+
"grad_norm": 0.6950928568840027,
|
931 |
+
"learning_rate": 0.00013842267553958371,
|
932 |
+
"loss": 0.0073,
|
933 |
+
"step": 1320
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 5.659574468085106,
|
937 |
+
"grad_norm": 0.05320321023464203,
|
938 |
+
"learning_rate": 0.00013759834061194794,
|
939 |
+
"loss": 0.0098,
|
940 |
+
"step": 1330
|
941 |
+
},
|
942 |
+
{
|
943 |
+
"epoch": 5.702127659574468,
|
944 |
+
"grad_norm": 0.17569933831691742,
|
945 |
+
"learning_rate": 0.00013677101928892554,
|
946 |
+
"loss": 0.0075,
|
947 |
+
"step": 1340
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 5.74468085106383,
|
951 |
+
"grad_norm": 0.052122730761766434,
|
952 |
+
"learning_rate": 0.00013594077728375128,
|
953 |
+
"loss": 0.0107,
|
954 |
+
"step": 1350
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 5.787234042553192,
|
958 |
+
"grad_norm": 0.2108752280473709,
|
959 |
+
"learning_rate": 0.00013510768054164653,
|
960 |
+
"loss": 0.0119,
|
961 |
+
"step": 1360
|
962 |
+
},
|
963 |
+
{
|
964 |
+
"epoch": 5.829787234042553,
|
965 |
+
"grad_norm": 0.047486983239650726,
|
966 |
+
"learning_rate": 0.00013427179523458127,
|
967 |
+
"loss": 0.0092,
|
968 |
+
"step": 1370
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 5.872340425531915,
|
972 |
+
"grad_norm": 0.043320391327142715,
|
973 |
+
"learning_rate": 0.0001334331877560182,
|
974 |
+
"loss": 0.0081,
|
975 |
+
"step": 1380
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 5.914893617021277,
|
979 |
+
"grad_norm": 0.09155077487230301,
|
980 |
+
"learning_rate": 0.00013259192471563912,
|
981 |
+
"loss": 0.0091,
|
982 |
+
"step": 1390
|
983 |
+
},
|
984 |
+
{
|
985 |
+
"epoch": 5.957446808510638,
|
986 |
+
"grad_norm": 0.049143675714731216,
|
987 |
+
"learning_rate": 0.00013174807293405428,
|
988 |
+
"loss": 0.0089,
|
989 |
+
"step": 1400
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 6.0,
|
993 |
+
"grad_norm": 0.07365540415048599,
|
994 |
+
"learning_rate": 0.00013090169943749476,
|
995 |
+
"loss": 0.0068,
|
996 |
+
"step": 1410
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 6.042553191489362,
|
1000 |
+
"grad_norm": 0.11437718570232391,
|
1001 |
+
"learning_rate": 0.00013005287145248878,
|
1002 |
+
"loss": 0.0064,
|
1003 |
+
"step": 1420
|
1004 |
+
},
|
1005 |
+
{
|
1006 |
+
"epoch": 6.085106382978723,
|
1007 |
+
"grad_norm": 0.06010650098323822,
|
1008 |
+
"learning_rate": 0.0001292016564005219,
|
1009 |
+
"loss": 0.0074,
|
1010 |
+
"step": 1430
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 6.127659574468085,
|
1014 |
+
"grad_norm": 0.04595167934894562,
|
1015 |
+
"learning_rate": 0.0001283481218926818,
|
1016 |
+
"loss": 0.0066,
|
1017 |
+
"step": 1440
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"epoch": 6.170212765957447,
|
1021 |
+
"grad_norm": 0.05351310595870018,
|
1022 |
+
"learning_rate": 0.00012749233572428804,
|
1023 |
+
"loss": 0.0097,
|
1024 |
+
"step": 1450
|
1025 |
+
},
|
1026 |
+
{
|
1027 |
+
"epoch": 6.212765957446808,
|
1028 |
+
"grad_norm": 0.13630953431129456,
|
1029 |
+
"learning_rate": 0.00012663436586950714,
|
1030 |
+
"loss": 0.0079,
|
1031 |
+
"step": 1460
|
1032 |
+
},
|
1033 |
+
{
|
1034 |
+
"epoch": 6.25531914893617,
|
1035 |
+
"grad_norm": 0.17622588574886322,
|
1036 |
+
"learning_rate": 0.00012577428047595344,
|
1037 |
+
"loss": 0.0084,
|
1038 |
+
"step": 1470
|
1039 |
+
},
|
1040 |
+
{
|
1041 |
+
"epoch": 6.297872340425532,
|
1042 |
+
"grad_norm": 0.050954531878232956,
|
1043 |
+
"learning_rate": 0.0001249121478592762,
|
1044 |
+
"loss": 0.0077,
|
1045 |
+
"step": 1480
|
1046 |
+
},
|
1047 |
+
{
|
1048 |
+
"epoch": 6.340425531914893,
|
1049 |
+
"grad_norm": 0.3051726818084717,
|
1050 |
+
"learning_rate": 0.0001240480364977335,
|
1051 |
+
"loss": 0.0085,
|
1052 |
+
"step": 1490
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 6.382978723404255,
|
1056 |
+
"grad_norm": 0.30509302020072937,
|
1057 |
+
"learning_rate": 0.00012318201502675285,
|
1058 |
+
"loss": 0.0092,
|
1059 |
+
"step": 1500
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 6.425531914893617,
|
1063 |
+
"grad_norm": 0.09164142608642578,
|
1064 |
+
"learning_rate": 0.00012231415223347972,
|
1065 |
+
"loss": 0.008,
|
1066 |
+
"step": 1510
|
1067 |
+
},
|
1068 |
+
{
|
1069 |
+
"epoch": 6.468085106382979,
|
1070 |
+
"grad_norm": 0.05406223237514496,
|
1071 |
+
"learning_rate": 0.0001214445170513139,
|
1072 |
+
"loss": 0.0078,
|
1073 |
+
"step": 1520
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 6.51063829787234,
|
1077 |
+
"grad_norm": 0.05845744535326958,
|
1078 |
+
"learning_rate": 0.00012057317855443395,
|
1079 |
+
"loss": 0.0092,
|
1080 |
+
"step": 1530
|
1081 |
+
},
|
1082 |
+
{
|
1083 |
+
"epoch": 6.553191489361702,
|
1084 |
+
"grad_norm": 0.05021122843027115,
|
1085 |
+
"learning_rate": 0.00011970020595231101,
|
1086 |
+
"loss": 0.007,
|
1087 |
+
"step": 1540
|
1088 |
+
},
|
1089 |
+
{
|
1090 |
+
"epoch": 6.595744680851064,
|
1091 |
+
"grad_norm": 0.10315235704183578,
|
1092 |
+
"learning_rate": 0.00011882566858421135,
|
1093 |
+
"loss": 0.0068,
|
1094 |
+
"step": 1550
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 6.638297872340425,
|
1098 |
+
"grad_norm": 0.08750782907009125,
|
1099 |
+
"learning_rate": 0.00011794963591368893,
|
1100 |
+
"loss": 0.009,
|
1101 |
+
"step": 1560
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 6.680851063829787,
|
1105 |
+
"grad_norm": 0.05412838235497475,
|
1106 |
+
"learning_rate": 0.0001170721775230679,
|
1107 |
+
"loss": 0.0071,
|
1108 |
+
"step": 1570
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 6.723404255319149,
|
1112 |
+
"grad_norm": 0.17292432487010956,
|
1113 |
+
"learning_rate": 0.00011619336310791586,
|
1114 |
+
"loss": 0.0091,
|
1115 |
+
"step": 1580
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 6.76595744680851,
|
1119 |
+
"grad_norm": 0.05503688380122185,
|
1120 |
+
"learning_rate": 0.00011531326247150803,
|
1121 |
+
"loss": 0.0069,
|
1122 |
+
"step": 1590
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 6.808510638297872,
|
1126 |
+
"grad_norm": 0.05121155083179474,
|
1127 |
+
"learning_rate": 0.00011443194551928266,
|
1128 |
+
"loss": 0.008,
|
1129 |
+
"step": 1600
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 6.851063829787234,
|
1133 |
+
"grad_norm": 0.0626005157828331,
|
1134 |
+
"learning_rate": 0.00011354948225328877,
|
1135 |
+
"loss": 0.0065,
|
1136 |
+
"step": 1610
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 6.8936170212765955,
|
1140 |
+
"grad_norm": 0.058921247720718384,
|
1141 |
+
"learning_rate": 0.0001126659427666257,
|
1142 |
+
"loss": 0.0078,
|
1143 |
+
"step": 1620
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 6.9361702127659575,
|
1147 |
+
"grad_norm": 0.058523211628198624,
|
1148 |
+
"learning_rate": 0.00011178139723787597,
|
1149 |
+
"loss": 0.008,
|
1150 |
+
"step": 1630
|
1151 |
+
},
|
1152 |
+
{
|
1153 |
+
"epoch": 6.9787234042553195,
|
1154 |
+
"grad_norm": 0.18594586849212646,
|
1155 |
+
"learning_rate": 0.00011089591592553082,
|
1156 |
+
"loss": 0.0076,
|
1157 |
+
"step": 1640
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 7.0212765957446805,
|
1161 |
+
"grad_norm": 0.053747180849313736,
|
1162 |
+
"learning_rate": 0.00011000956916240985,
|
1163 |
+
"loss": 0.0074,
|
1164 |
+
"step": 1650
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 7.0638297872340425,
|
1168 |
+
"grad_norm": 0.03964696079492569,
|
1169 |
+
"learning_rate": 0.00010912242735007441,
|
1170 |
+
"loss": 0.0071,
|
1171 |
+
"step": 1660
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 7.1063829787234045,
|
1175 |
+
"grad_norm": 0.05952566862106323,
|
1176 |
+
"learning_rate": 0.00010823456095323579,
|
1177 |
+
"loss": 0.0065,
|
1178 |
+
"step": 1670
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 7.148936170212766,
|
1182 |
+
"grad_norm": 0.21424749493598938,
|
1183 |
+
"learning_rate": 0.00010734604049415822,
|
1184 |
+
"loss": 0.0075,
|
1185 |
+
"step": 1680
|
1186 |
+
},
|
1187 |
+
{
|
1188 |
+
"epoch": 7.191489361702128,
|
1189 |
+
"grad_norm": 0.03922433406114578,
|
1190 |
+
"learning_rate": 0.0001064569365470574,
|
1191 |
+
"loss": 0.0071,
|
1192 |
+
"step": 1690
|
1193 |
+
},
|
1194 |
+
{
|
1195 |
+
"epoch": 7.23404255319149,
|
1196 |
+
"grad_norm": 0.05505215749144554,
|
1197 |
+
"learning_rate": 0.00010556731973249485,
|
1198 |
+
"loss": 0.0061,
|
1199 |
+
"step": 1700
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 7.276595744680851,
|
1203 |
+
"grad_norm": 0.03699268028140068,
|
1204 |
+
"learning_rate": 0.00010467726071176853,
|
1205 |
+
"loss": 0.0075,
|
1206 |
+
"step": 1710
|
1207 |
+
},
|
1208 |
+
{
|
1209 |
+
"epoch": 7.319148936170213,
|
1210 |
+
"grad_norm": 0.04546520113945007,
|
1211 |
+
"learning_rate": 0.00010378683018130047,
|
1212 |
+
"loss": 0.0072,
|
1213 |
+
"step": 1720
|
1214 |
+
},
|
1215 |
+
{
|
1216 |
+
"epoch": 7.361702127659575,
|
1217 |
+
"grad_norm": 0.056984953582286835,
|
1218 |
+
"learning_rate": 0.0001028960988670212,
|
1219 |
+
"loss": 0.007,
|
1220 |
+
"step": 1730
|
1221 |
+
},
|
1222 |
+
{
|
1223 |
+
"epoch": 7.404255319148936,
|
1224 |
+
"grad_norm": 0.24714773893356323,
|
1225 |
+
"learning_rate": 0.00010200513751875227,
|
1226 |
+
"loss": 0.0074,
|
1227 |
+
"step": 1740
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 7.446808510638298,
|
1231 |
+
"grad_norm": 0.06558862328529358,
|
1232 |
+
"learning_rate": 0.00010111401690458654,
|
1233 |
+
"loss": 0.0064,
|
1234 |
+
"step": 1750
|
1235 |
+
},
|
1236 |
+
{
|
1237 |
+
"epoch": 7.48936170212766,
|
1238 |
+
"grad_norm": 0.06254340708255768,
|
1239 |
+
"learning_rate": 0.00010022280780526725,
|
1240 |
+
"loss": 0.0076,
|
1241 |
+
"step": 1760
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 7.531914893617021,
|
1245 |
+
"grad_norm": 0.08606445789337158,
|
1246 |
+
"learning_rate": 9.93315810085658e-05,
|
1247 |
+
"loss": 0.0076,
|
1248 |
+
"step": 1770
|
1249 |
+
},
|
1250 |
+
{
|
1251 |
+
"epoch": 7.574468085106383,
|
1252 |
+
"grad_norm": 0.051956657320261,
|
1253 |
+
"learning_rate": 9.844040730365936e-05,
|
1254 |
+
"loss": 0.0073,
|
1255 |
+
"step": 1780
|
1256 |
+
},
|
1257 |
+
{
|
1258 |
+
"epoch": 7.617021276595745,
|
1259 |
+
"grad_norm": 0.24819940328598022,
|
1260 |
+
"learning_rate": 9.754935747550804e-05,
|
1261 |
+
"loss": 0.0077,
|
1262 |
+
"step": 1790
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 7.659574468085106,
|
1266 |
+
"grad_norm": 0.039045918732881546,
|
1267 |
+
"learning_rate": 9.665850229923258e-05,
|
1268 |
+
"loss": 0.0071,
|
1269 |
+
"step": 1800
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 7.702127659574468,
|
1273 |
+
"grad_norm": 0.04438428208231926,
|
1274 |
+
"learning_rate": 9.57679125344927e-05,
|
1275 |
+
"loss": 0.0074,
|
1276 |
+
"step": 1810
|
1277 |
+
},
|
1278 |
+
{
|
1279 |
+
"epoch": 7.74468085106383,
|
1280 |
+
"grad_norm": 0.31067439913749695,
|
1281 |
+
"learning_rate": 9.487765891986682e-05,
|
1282 |
+
"loss": 0.0087,
|
1283 |
+
"step": 1820
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 7.787234042553192,
|
1287 |
+
"grad_norm": 0.06251853704452515,
|
1288 |
+
"learning_rate": 9.398781216723331e-05,
|
1289 |
+
"loss": 0.0069,
|
1290 |
+
"step": 1830
|
1291 |
+
},
|
1292 |
+
{
|
1293 |
+
"epoch": 7.829787234042553,
|
1294 |
+
"grad_norm": 0.04706185683608055,
|
1295 |
+
"learning_rate": 9.309844295615389e-05,
|
1296 |
+
"loss": 0.0072,
|
1297 |
+
"step": 1840
|
1298 |
+
},
|
1299 |
+
{
|
1300 |
+
"epoch": 7.872340425531915,
|
1301 |
+
"grad_norm": 0.03839968144893646,
|
1302 |
+
"learning_rate": 9.220962192825968e-05,
|
1303 |
+
"loss": 0.0069,
|
1304 |
+
"step": 1850
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 7.914893617021277,
|
1308 |
+
"grad_norm": 0.03722318261861801,
|
1309 |
+
"learning_rate": 9.132141968164026e-05,
|
1310 |
+
"loss": 0.0069,
|
1311 |
+
"step": 1860
|
1312 |
+
},
|
1313 |
+
{
|
1314 |
+
"epoch": 7.957446808510638,
|
1315 |
+
"grad_norm": 0.0543668158352375,
|
1316 |
+
"learning_rate": 9.043390676523604e-05,
|
1317 |
+
"loss": 0.0076,
|
1318 |
+
"step": 1870
|
1319 |
+
},
|
1320 |
+
{
|
1321 |
+
"epoch": 8.0,
|
1322 |
+
"grad_norm": 0.051696889102458954,
|
1323 |
+
"learning_rate": 8.954715367323468e-05,
|
1324 |
+
"loss": 0.0066,
|
1325 |
+
"step": 1880
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 8.042553191489361,
|
1329 |
+
"grad_norm": 0.08786992728710175,
|
1330 |
+
"learning_rate": 8.866123083947182e-05,
|
1331 |
+
"loss": 0.0062,
|
1332 |
+
"step": 1890
|
1333 |
+
},
|
1334 |
+
{
|
1335 |
+
"epoch": 8.085106382978724,
|
1336 |
+
"grad_norm": 0.3556790351867676,
|
1337 |
+
"learning_rate": 8.777620863183657e-05,
|
1338 |
+
"loss": 0.0079,
|
1339 |
+
"step": 1900
|
1340 |
+
},
|
1341 |
+
{
|
1342 |
+
"epoch": 8.127659574468085,
|
1343 |
+
"grad_norm": 0.08049122244119644,
|
1344 |
+
"learning_rate": 8.689215734668232e-05,
|
1345 |
+
"loss": 0.0064,
|
1346 |
+
"step": 1910
|
1347 |
+
},
|
1348 |
+
{
|
1349 |
+
"epoch": 8.170212765957446,
|
1350 |
+
"grad_norm": 0.05408492311835289,
|
1351 |
+
"learning_rate": 8.600914720324316e-05,
|
1352 |
+
"loss": 0.0077,
|
1353 |
+
"step": 1920
|
1354 |
+
},
|
1355 |
+
{
|
1356 |
+
"epoch": 8.212765957446809,
|
1357 |
+
"grad_norm": 0.034946855157613754,
|
1358 |
+
"learning_rate": 8.512724833805634e-05,
|
1359 |
+
"loss": 0.0067,
|
1360 |
+
"step": 1930
|
1361 |
+
},
|
1362 |
+
{
|
1363 |
+
"epoch": 8.25531914893617,
|
1364 |
+
"grad_norm": 0.05422681197524071,
|
1365 |
+
"learning_rate": 8.424653079939156e-05,
|
1366 |
+
"loss": 0.0062,
|
1367 |
+
"step": 1940
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 8.297872340425531,
|
1371 |
+
"grad_norm": 0.06405791640281677,
|
1372 |
+
"learning_rate": 8.336706454168701e-05,
|
1373 |
+
"loss": 0.0064,
|
1374 |
+
"step": 1950
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 8.340425531914894,
|
1378 |
+
"grad_norm": 0.08694509416818619,
|
1379 |
+
"learning_rate": 8.248891941999297e-05,
|
1380 |
+
"loss": 0.006,
|
1381 |
+
"step": 1960
|
1382 |
+
},
|
1383 |
+
{
|
1384 |
+
"epoch": 8.382978723404255,
|
1385 |
+
"grad_norm": 0.05589594319462776,
|
1386 |
+
"learning_rate": 8.161216518442334e-05,
|
1387 |
+
"loss": 0.0067,
|
1388 |
+
"step": 1970
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 8.425531914893616,
|
1392 |
+
"grad_norm": 0.05914654955267906,
|
1393 |
+
"learning_rate": 8.073687147461547e-05,
|
1394 |
+
"loss": 0.0065,
|
1395 |
+
"step": 1980
|
1396 |
+
},
|
1397 |
+
{
|
1398 |
+
"epoch": 8.46808510638298,
|
1399 |
+
"grad_norm": 0.05237606540322304,
|
1400 |
+
"learning_rate": 7.98631078141987e-05,
|
1401 |
+
"loss": 0.0071,
|
1402 |
+
"step": 1990
|
1403 |
+
},
|
1404 |
+
{
|
1405 |
+
"epoch": 8.51063829787234,
|
1406 |
+
"grad_norm": 0.056138359010219574,
|
1407 |
+
"learning_rate": 7.89909436052722e-05,
|
1408 |
+
"loss": 0.0079,
|
1409 |
+
"step": 2000
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 8.553191489361701,
|
1413 |
+
"grad_norm": 0.06285829097032547,
|
1414 |
+
"learning_rate": 7.812044812289249e-05,
|
1415 |
+
"loss": 0.0064,
|
1416 |
+
"step": 2010
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 8.595744680851064,
|
1420 |
+
"grad_norm": 0.04014954715967178,
|
1421 |
+
"learning_rate": 7.72516905095709e-05,
|
1422 |
+
"loss": 0.0072,
|
1423 |
+
"step": 2020
|
1424 |
+
},
|
1425 |
+
{
|
1426 |
+
"epoch": 8.638297872340425,
|
1427 |
+
"grad_norm": 0.04423344135284424,
|
1428 |
+
"learning_rate": 7.638473976978177e-05,
|
1429 |
+
"loss": 0.0065,
|
1430 |
+
"step": 2030
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 8.680851063829786,
|
1434 |
+
"grad_norm": 0.05287999287247658,
|
1435 |
+
"learning_rate": 7.55196647644814e-05,
|
1436 |
+
"loss": 0.0081,
|
1437 |
+
"step": 2040
|
1438 |
+
},
|
1439 |
+
{
|
1440 |
+
"epoch": 8.72340425531915,
|
1441 |
+
"grad_norm": 0.056493304669857025,
|
1442 |
+
"learning_rate": 7.465653420563845e-05,
|
1443 |
+
"loss": 0.0067,
|
1444 |
+
"step": 2050
|
1445 |
+
},
|
1446 |
+
{
|
1447 |
+
"epoch": 8.76595744680851,
|
1448 |
+
"grad_norm": 0.057281821966171265,
|
1449 |
+
"learning_rate": 7.379541665077643e-05,
|
1450 |
+
"loss": 0.0078,
|
1451 |
+
"step": 2060
|
1452 |
+
},
|
1453 |
+
{
|
1454 |
+
"epoch": 8.808510638297872,
|
1455 |
+
"grad_norm": 0.053731102496385574,
|
1456 |
+
"learning_rate": 7.293638049752812e-05,
|
1457 |
+
"loss": 0.0066,
|
1458 |
+
"step": 2070
|
1459 |
+
},
|
1460 |
+
{
|
1461 |
+
"epoch": 8.851063829787234,
|
1462 |
+
"grad_norm": 0.18697482347488403,
|
1463 |
+
"learning_rate": 7.207949397820278e-05,
|
1464 |
+
"loss": 0.0069,
|
1465 |
+
"step": 2080
|
1466 |
+
},
|
1467 |
+
{
|
1468 |
+
"epoch": 8.893617021276595,
|
1469 |
+
"grad_norm": 0.0387713797390461,
|
1470 |
+
"learning_rate": 7.122482515436661e-05,
|
1471 |
+
"loss": 0.007,
|
1472 |
+
"step": 2090
|
1473 |
+
},
|
1474 |
+
{
|
1475 |
+
"epoch": 8.936170212765958,
|
1476 |
+
"grad_norm": 0.04441935196518898,
|
1477 |
+
"learning_rate": 7.037244191143661e-05,
|
1478 |
+
"loss": 0.0067,
|
1479 |
+
"step": 2100
|
1480 |
+
},
|
1481 |
+
{
|
1482 |
+
"epoch": 8.97872340425532,
|
1483 |
+
"grad_norm": 0.05758345127105713,
|
1484 |
+
"learning_rate": 6.952241195328868e-05,
|
1485 |
+
"loss": 0.0065,
|
1486 |
+
"step": 2110
|
1487 |
+
},
|
1488 |
+
{
|
1489 |
+
"epoch": 9.02127659574468,
|
1490 |
+
"grad_norm": 0.050706226378679276,
|
1491 |
+
"learning_rate": 6.867480279687974e-05,
|
1492 |
+
"loss": 0.0063,
|
1493 |
+
"step": 2120
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 9.063829787234043,
|
1497 |
+
"grad_norm": 0.05180887505412102,
|
1498 |
+
"learning_rate": 6.782968176688514e-05,
|
1499 |
+
"loss": 0.0062,
|
1500 |
+
"step": 2130
|
1501 |
+
},
|
1502 |
+
{
|
1503 |
+
"epoch": 9.106382978723405,
|
1504 |
+
"grad_norm": 0.05492401868104935,
|
1505 |
+
"learning_rate": 6.6987115990351e-05,
|
1506 |
+
"loss": 0.006,
|
1507 |
+
"step": 2140
|
1508 |
+
},
|
1509 |
+
{
|
1510 |
+
"epoch": 9.148936170212766,
|
1511 |
+
"grad_norm": 0.053439777344465256,
|
1512 |
+
"learning_rate": 6.614717239136246e-05,
|
1513 |
+
"loss": 0.0066,
|
1514 |
+
"step": 2150
|
1515 |
+
},
|
1516 |
+
{
|
1517 |
+
"epoch": 9.191489361702128,
|
1518 |
+
"grad_norm": 0.15650008618831635,
|
1519 |
+
"learning_rate": 6.530991768572794e-05,
|
1520 |
+
"loss": 0.006,
|
1521 |
+
"step": 2160
|
1522 |
+
},
|
1523 |
+
{
|
1524 |
+
"epoch": 9.23404255319149,
|
1525 |
+
"grad_norm": 0.04846300184726715,
|
1526 |
+
"learning_rate": 6.447541837568e-05,
|
1527 |
+
"loss": 0.0068,
|
1528 |
+
"step": 2170
|
1529 |
+
},
|
1530 |
+
{
|
1531 |
+
"epoch": 9.27659574468085,
|
1532 |
+
"grad_norm": 0.047157011926174164,
|
1533 |
+
"learning_rate": 6.364374074459307e-05,
|
1534 |
+
"loss": 0.006,
|
1535 |
+
"step": 2180
|
1536 |
+
},
|
1537 |
+
{
|
1538 |
+
"epoch": 9.319148936170214,
|
1539 |
+
"grad_norm": 0.047358132898807526,
|
1540 |
+
"learning_rate": 6.281495085171869e-05,
|
1541 |
+
"loss": 0.0058,
|
1542 |
+
"step": 2190
|
1543 |
+
},
|
1544 |
+
{
|
1545 |
+
"epoch": 9.361702127659575,
|
1546 |
+
"grad_norm": 0.11200093477964401,
|
1547 |
+
"learning_rate": 6.198911452693853e-05,
|
1548 |
+
"loss": 0.007,
|
1549 |
+
"step": 2200
|
1550 |
+
},
|
1551 |
+
{
|
1552 |
+
"epoch": 9.404255319148936,
|
1553 |
+
"grad_norm": 0.06481984257698059,
|
1554 |
+
"learning_rate": 6.116629736553552e-05,
|
1555 |
+
"loss": 0.0069,
|
1556 |
+
"step": 2210
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 9.446808510638299,
|
1560 |
+
"grad_norm": 0.04743931442499161,
|
1561 |
+
"learning_rate": 6.0346564722983736e-05,
|
1562 |
+
"loss": 0.0072,
|
1563 |
+
"step": 2220
|
1564 |
+
},
|
1565 |
+
{
|
1566 |
+
"epoch": 9.48936170212766,
|
1567 |
+
"grad_norm": 0.07307924330234528,
|
1568 |
+
"learning_rate": 5.952998170975724e-05,
|
1569 |
+
"loss": 0.0062,
|
1570 |
+
"step": 2230
|
1571 |
+
},
|
1572 |
+
{
|
1573 |
+
"epoch": 9.53191489361702,
|
1574 |
+
"grad_norm": 0.05650079995393753,
|
1575 |
+
"learning_rate": 5.871661318615848e-05,
|
1576 |
+
"loss": 0.0061,
|
1577 |
+
"step": 2240
|
1578 |
+
},
|
1579 |
+
{
|
1580 |
+
"epoch": 9.574468085106384,
|
1581 |
+
"grad_norm": 0.057734813541173935,
|
1582 |
+
"learning_rate": 5.790652375716652e-05,
|
1583 |
+
"loss": 0.0068,
|
1584 |
+
"step": 2250
|
1585 |
+
},
|
1586 |
+
{
|
1587 |
+
"epoch": 9.617021276595745,
|
1588 |
+
"grad_norm": 0.05608903244137764,
|
1589 |
+
"learning_rate": 5.709977776730537e-05,
|
1590 |
+
"loss": 0.0071,
|
1591 |
+
"step": 2260
|
1592 |
+
},
|
1593 |
+
{
|
1594 |
+
"epoch": 9.659574468085106,
|
1595 |
+
"grad_norm": 0.07133158296346664,
|
1596 |
+
"learning_rate": 5.62964392955335e-05,
|
1597 |
+
"loss": 0.0067,
|
1598 |
+
"step": 2270
|
1599 |
+
},
|
1600 |
+
{
|
1601 |
+
"epoch": 9.702127659574469,
|
1602 |
+
"grad_norm": 0.2167043685913086,
|
1603 |
+
"learning_rate": 5.549657215015367e-05,
|
1604 |
+
"loss": 0.0067,
|
1605 |
+
"step": 2280
|
1606 |
+
},
|
1607 |
+
{
|
1608 |
+
"epoch": 9.74468085106383,
|
1609 |
+
"grad_norm": 0.06523007154464722,
|
1610 |
+
"learning_rate": 5.470023986374516e-05,
|
1611 |
+
"loss": 0.0068,
|
1612 |
+
"step": 2290
|
1613 |
+
},
|
1614 |
+
{
|
1615 |
+
"epoch": 9.787234042553191,
|
1616 |
+
"grad_norm": 0.04895370826125145,
|
1617 |
+
"learning_rate": 5.39075056881172e-05,
|
1618 |
+
"loss": 0.0058,
|
1619 |
+
"step": 2300
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 9.829787234042554,
|
1623 |
+
"grad_norm": 0.056433409452438354,
|
1624 |
+
"learning_rate": 5.31184325892849e-05,
|
1625 |
+
"loss": 0.0067,
|
1626 |
+
"step": 2310
|
1627 |
+
},
|
1628 |
+
{
|
1629 |
+
"epoch": 9.872340425531915,
|
1630 |
+
"grad_norm": 0.049970466643571854,
|
1631 |
+
"learning_rate": 5.233308324246805e-05,
|
1632 |
+
"loss": 0.0058,
|
1633 |
+
"step": 2320
|
1634 |
+
},
|
1635 |
+
{
|
1636 |
+
"epoch": 9.914893617021276,
|
1637 |
+
"grad_norm": 0.08367093652486801,
|
1638 |
+
"learning_rate": 5.155152002711285e-05,
|
1639 |
+
"loss": 0.0063,
|
1640 |
+
"step": 2330
|
1641 |
+
},
|
1642 |
+
{
|
1643 |
+
"epoch": 9.957446808510639,
|
1644 |
+
"grad_norm": 0.07821632921695709,
|
1645 |
+
"learning_rate": 5.077380502193725e-05,
|
1646 |
+
"loss": 0.0071,
|
1647 |
+
"step": 2340
|
1648 |
+
},
|
1649 |
+
{
|
1650 |
+
"epoch": 10.0,
|
1651 |
+
"grad_norm": 0.04344159737229347,
|
1652 |
+
"learning_rate": 5.000000000000002e-05,
|
1653 |
+
"loss": 0.0078,
|
1654 |
+
"step": 2350
|
1655 |
}
|
1656 |
],
|
1657 |
"logging_steps": 10,
|
|
|
1671 |
"attributes": {}
|
1672 |
}
|
1673 |
},
|
1674 |
+
"total_flos": 9.44675050315776e+16,
|
1675 |
"train_batch_size": 1,
|
1676 |
"trial_name": null,
|
1677 |
"trial_params": null
|