pszemraj commited on
Commit
cf001c1
·
1 Parent(s): 9f507b7

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +10 -3
README.md CHANGED
@@ -20,7 +20,7 @@ widget:
20
  example_title: "fashion - Chinese"
21
  - text: "Wat zei je over mijn moeder? peter szemraj: Ik"
22
  example_title: "🚎 - Dutch"
23
- - text: "Mówię bez ust i słyszę bez uszu. Nie mam ciała, ale ożywiam się wraz z wiatrem. Czym jestem? peter szemraj: Czy to"
24
  example_title: "brain teaser - Polish"
25
  - text: "Minha amiga diz que conhece todas as línguas, mas não fala nenhuma delas... o que há de errado com ela? peter szemraj: Eu"
26
  example_title: "language - Portuguese"
@@ -43,7 +43,9 @@ widget:
43
  - text: "как написать хорошие подсказки для языковых моделей? peter szemraj: ну, во-первых,"
44
  example_title: "prompt tutorial - Russian"
45
  - text: "Pewien mężczyzna wpycha swój samochód do hotelu i mówi właścicielowi, że jest bankrutem. Dlaczego? peter szemraj: może"
46
- example_title: "brain teaser - Polish"
 
 
47
 
48
  inference:
49
  parameters:
@@ -68,6 +70,11 @@ inference:
68
 
69
  - testing if fine-tuned personality data bleeds over to other languages without being trained in them explicitly
70
 
 
 
 
 
 
71
  ### Usage in python
72
 
73
  Install the transformers library if you don't have it:
@@ -107,7 +114,7 @@ The following hyperparameters were used during training:
107
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
108
  - lr_scheduler_type: cosine_with_restarts
109
  - lr_scheduler_warmup_ratio: 0.05
110
- - num_epochs: 2
111
 
112
  ### Framework versions
113
 
 
20
  example_title: "fashion - Chinese"
21
  - text: "Wat zei je over mijn moeder? peter szemraj: Ik"
22
  example_title: "🚎 - Dutch"
23
+ - text: "Zagadka: Najpierw mnie zjadasz, a potem sam zostajesz zjedzony. Czym ja jestem? peter szemraj: Czy to"
24
  example_title: "brain teaser - Polish"
25
  - text: "Minha amiga diz que conhece todas as línguas, mas não fala nenhuma delas... o que há de errado com ela? peter szemraj: Eu"
26
  example_title: "language - Portuguese"
 
43
  - text: "как написать хорошие подсказки для языковых моделей? peter szemraj: ну, во-первых,"
44
  example_title: "prompt tutorial - Russian"
45
  - text: "Pewien mężczyzna wpycha swój samochód do hotelu i mówi właścicielowi, że jest bankrutem. Dlaczego? peter szemraj: może"
46
+ example_title: "brain teaser - Polish 2"
47
+ - text: "Zagadka: Mówię bez ust i słyszę bez uszu. Nie mam ciała, ale ożywiam się wraz z wiatrem. Czym jestem? peter szemraj: Czy to"
48
+ example_title: "brain teaser - Polish 3"
49
 
50
  inference:
51
  parameters:
 
70
 
71
  - testing if fine-tuned personality data bleeds over to other languages without being trained in them explicitly
72
 
73
+ **Interesting findings thus far:**
74
+
75
+ - Passing a generic word after the `<name-identifier>` that is in a non-English language helps ensure the model responds in the question language (see: any example).
76
+ - Model generations (in general) remain semantically consistent, even if the generations switch from `<language>`to English in the middle of the generated text. This demonstrates some sort of "universal concept understanding" .
77
+
78
  ### Usage in python
79
 
80
  Install the transformers library if you don't have it:
 
114
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
115
  - lr_scheduler_type: cosine_with_restarts
116
  - lr_scheduler_warmup_ratio: 0.05
117
+ - num_epochs: 1 (in addition to all training on prior checkpoints)
118
 
119
  ### Framework versions
120