Views
No views yet
oscar_unshuffled_deduplicated_de (excluding validation set; original model was trained for 75 epochs on less data)1>>> from transformers import pipeline, set_seed
2>>> generator = pipeline('text-generation', model='malteos/gpt2-xl-wechsel-german')
3>>> set_seed(42)
4>>> generator("Hello, I'm a language model,", max_length=30, num_return_sequences=5)
5
6[{'generated_text': "Hello, I'm a language model, a language for thinking, a language for expressing thoughts."},
7 {'generated_text': "Hello, I'm a language model, a compiler, a compiler library, I just want to know how I build this kind of stuff. I don"},
8 {'generated_text': "Hello, I'm a language model, and also have more than a few of your own, but I understand that they're going to need some help"},
9 {'generated_text': "Hello, I'm a language model, a system model. I want to know my language so that it might be more interesting, more user-friendly"},
10 {'generated_text': 'Hello, I\'m a language model, not a language model"\n\nThe concept of "no-tricks" comes in handy later with new'}]1from transformers import GPT2Tokenizer, GPT2Model
2tokenizer = GPT2Tokenizer.from_pretrained('malteos/gpt2-xl-wechsel-german')
3model = GPT2Model.from_pretrained('malteos/gpt2-xl-wechsel-german')
4text = "Replace me by any text you'd like."
5encoded_input = tokenizer(text, return_tensors='pt')
6output = model(**encoded_input)| Model (size) | PPL |
|---|---|
gpt2-xl-wechsel-german (1.5B) | 14.5 |
gpt2-wechsel-german-ds-meg (117M) | 26.4 |
gpt2-wechsel-german (117M) | 26.8 |
gpt2 (retrained from scratch) (117M) | 27.63 |