README.md
Browse files
README.md
CHANGED
|
@@ -6,3 +6,32 @@ language:
|
|
| 6 |
---
|
| 7 |
|
| 8 |
# German Tinyllama-120M
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
---
|
| 7 |
|
| 8 |
# German Tinyllama-120M
|
| 9 |
+
|
| 10 |
+
This is a German Tinyllama 120M language model trained from scratch using the
|
| 11 |
+
the [Tinyllama](https://github.com/jzhang38/TinyLlama) codebase on the German portion of [RedPajama V2](https://huggingface.co/datasets/togethercomputer/RedPajama-Data-V2).
|
| 12 |
+
|
| 13 |
+
|
| 14 |
+
### Usage
|
| 15 |
+
|
| 16 |
+
```python
|
| 17 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
| 18 |
+
|
| 19 |
+
model = AutoModelForCausalLM.from_pretrained("LSX-UniWue/german_tinyllama_120M")
|
| 20 |
+
|
| 21 |
+
tokenizer = AutoTokenizer.from_pretrained("LSX-UniWue/german_tinyllama_120M")
|
| 22 |
+
```
|
| 23 |
+
|
| 24 |
+
|
| 25 |
+
### Performance
|
| 26 |
+
We evaluated our results on the [SuperGLEBer](https://lsx-uniwue.github.io/SuperGLEBer-site/) benchmark.
|
| 27 |
+
|
| 28 |
+
| Task Type | Task Name | Metric | Score |
|
| 29 |
+
|---------------------|--------------|----------|-------|
|
| 30 |
+
| Classification | NLI | Accuracy | 0.629 |
|
| 31 |
+
| Classification | DB Aspect | micro F1 | 0.517 |
|
| 32 |
+
| Sequence Tagging | NER Europarl | micro F1 | 0.538 |
|
| 33 |
+
| Sentence Similarity | Pawsx | Pearson | 0.489 |
|
| 34 |
+
| Question Answering | MLQA | F1 | 0.846 |
|
| 35 |
+
|
| 36 |
+
|
| 37 |
+
|