Update README.md
Browse files
README.md
CHANGED
|
@@ -53,17 +53,15 @@ which is the original mC4, except
|
|
| 53 |
|
| 54 |
TL;DR: [yhavinga/gpt2-medium-dutch](https://huggingface.co/yhavinga/gpt2-medium-dutch) is the best model.
|
| 55 |
|
| 56 |
-
* `yhavinga/gpt-neo-125M-dutch` is trained on a fraction of C4 containing only wikipedia and news sites.
|
| 57 |
* The models with `a`/`b` in the step-column have been trained to step `a` of a total of `b` steps.
|
| 58 |
|
| 59 |
| | model | params | train seq len | ppl | loss | batch size | epochs | steps | optim | lr | duration | config |
|
| 60 |
|-----------------------------------------------------------------------------------|---------|--------|---------------|------|------|------------|--------|-----------------|-----------|--------|----------|-----------|
|
| 61 |
-
| [yhavinga/gpt-neo-125M-dutch](https://huggingface.co/yhavinga/gpt-neo-125M-dutch) | gpt neo | 125M | 512 |
|
| 62 |
| [yhavinga/gpt2-medium-dutch](https://huggingface.co/yhavinga/gpt2-medium-dutch) | gpt2 | 345M | 512 | 15.1 | 2.71 | 128 | 4 | 320000/520502 | adafactor | 8e-4 | 7d 2h | full |
|
| 63 |
| [yhavinga/gpt2-large-dutch](https://huggingface.co/yhavinga/gpt2-large-dutch) | gpt2 | 762M | 512 | 15.1 | 2.72 | 32 | 1 | 1100000/2082009 | adafactor | 3.3e-5 | 8d 15h | large |
|
| 64 |
| [yhavinga/gpt-neo-1.3B-dutch](https://huggingface.co/yhavinga/gpt-neo-1.3B-dutch) | gpt neo | 1.3B | 512 | 16.0 | 2.77 | 16 | 1 | 960000/3049896 | adafactor | 5e-4 | 7d 11h | full |
|
| 65 |
|
| 66 |
-
|
| 67 |
## Acknowledgements
|
| 68 |
|
| 69 |
This project would not have been possible without compute generously provided by Google through the
|
|
|
|
| 53 |
|
| 54 |
TL;DR: [yhavinga/gpt2-medium-dutch](https://huggingface.co/yhavinga/gpt2-medium-dutch) is the best model.
|
| 55 |
|
|
|
|
| 56 |
* The models with `a`/`b` in the step-column have been trained to step `a` of a total of `b` steps.
|
| 57 |
|
| 58 |
| | model | params | train seq len | ppl | loss | batch size | epochs | steps | optim | lr | duration | config |
|
| 59 |
|-----------------------------------------------------------------------------------|---------|--------|---------------|------|------|------------|--------|-----------------|-----------|--------|----------|-----------|
|
| 60 |
+
| [yhavinga/gpt-neo-125M-dutch](https://huggingface.co/yhavinga/gpt-neo-125M-dutch) | gpt neo | 125M | 512 | 20.9 | 3.04 | 128 | 1 | 190000/558608 | adafactor | 2.4e-3 | 1d 12h | full |
|
| 61 |
| [yhavinga/gpt2-medium-dutch](https://huggingface.co/yhavinga/gpt2-medium-dutch) | gpt2 | 345M | 512 | 15.1 | 2.71 | 128 | 4 | 320000/520502 | adafactor | 8e-4 | 7d 2h | full |
|
| 62 |
| [yhavinga/gpt2-large-dutch](https://huggingface.co/yhavinga/gpt2-large-dutch) | gpt2 | 762M | 512 | 15.1 | 2.72 | 32 | 1 | 1100000/2082009 | adafactor | 3.3e-5 | 8d 15h | large |
|
| 63 |
| [yhavinga/gpt-neo-1.3B-dutch](https://huggingface.co/yhavinga/gpt-neo-1.3B-dutch) | gpt neo | 1.3B | 512 | 16.0 | 2.77 | 16 | 1 | 960000/3049896 | adafactor | 5e-4 | 7d 11h | full |
|
| 64 |
|
|
|
|
| 65 |
## Acknowledgements
|
| 66 |
|
| 67 |
This project would not have been possible without compute generously provided by Google through the
|