Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,33 @@
|
|
1 |
-
---
|
2 |
-
license: apache-2.0
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
base_model: google/t5-efficient-mini-nl24
|
4 |
+
datasets:
|
5 |
+
- Open-Orca/FLAN
|
6 |
+
language:
|
7 |
+
- en
|
8 |
+
library_name: transformers
|
9 |
+
pipeline_tag: text2text-generation
|
10 |
+
---
|
11 |
+
|
12 |
+
|
13 |
+
# t5e-mini-nl24-flan
|
14 |
+
|
15 |
+
|
16 |
+
25k steps on FLAN as an initial test/validation that [code](https://github.com/pszemraj/nanoT5/tree/flan-dataset) works. Not practically useful.
|
17 |
+
|
18 |
+
```py
|
19 |
+
from transformers import pipeline
|
20 |
+
|
21 |
+
pipe = pipeline(
|
22 |
+
"text2text-generation",
|
23 |
+
model="pszemraj/t5e-mini-nl24-flan",
|
24 |
+
)
|
25 |
+
res = pipe(
|
26 |
+
"How can mirrors be real if our eyes aren't real?",
|
27 |
+
top_k=4,
|
28 |
+
penalty_alpha=0.6,
|
29 |
+
max_new_tokens=128,
|
30 |
+
)
|
31 |
+
print(res[0]["generated_text"])
|
32 |
+
```
|
33 |
+
|