witcher
commited on
Commit
•
6ddf34d
1
Parent(s):
ae0fc7b
Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
## MechDistilGPT2
|
2 |
This model is fine-tuned on 200k text scraped from Mechanical/Automotive pdf books.
|
3 |
|
@@ -6,7 +17,6 @@ Base model is DistilGPT2(https://huggingface.co/gpt2) (the smallest version of G
|
|
6 |
## Fine-Tuning
|
7 |
* Default Training Args
|
8 |
* Epochs = 3
|
9 |
-
* Perplexity = 48
|
10 |
* Training set = 200k sentences
|
11 |
* Validation set = 40k sentences
|
12 |
|
@@ -14,4 +24,4 @@ Base model is DistilGPT2(https://huggingface.co/gpt2) (the smallest version of G
|
|
14 |
* Transformers 4.7.0.dev0
|
15 |
* Pytorch 1.8.1+cu111
|
16 |
* Datasets 1.6.2
|
17 |
-
* Tokenizers 0.10.2
|
|
|
1 |
+
|
2 |
+
---
|
3 |
+
tags:
|
4 |
+
- Causal Language modeling
|
5 |
+
model_index:
|
6 |
+
- name: MechDistilGPT2
|
7 |
+
results:
|
8 |
+
- task:
|
9 |
+
name: Causal Language modeling
|
10 |
+
type: Causal Language modeling
|
11 |
+
---
|
12 |
## MechDistilGPT2
|
13 |
This model is fine-tuned on 200k text scraped from Mechanical/Automotive pdf books.
|
14 |
|
|
|
17 |
## Fine-Tuning
|
18 |
* Default Training Args
|
19 |
* Epochs = 3
|
|
|
20 |
* Training set = 200k sentences
|
21 |
* Validation set = 40k sentences
|
22 |
|
|
|
24 |
* Transformers 4.7.0.dev0
|
25 |
* Pytorch 1.8.1+cu111
|
26 |
* Datasets 1.6.2
|
27 |
+
* Tokenizers 0.10.2
|