Downloads · 30 days
16
14% of all-time downloads
dmitry315/ELlama1-0.7b
ELlama1-0.7b is a text generation model from dmitry315. Use it when you need the model to write or continue text. The card lists the license as mit.
Downloads · 30 days
16
14% of all-time downloads
All-time downloads
112
Public
Parameters
748M
1.5 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors1.5 GB · 99%
From the Hugging Face model README
Серия LLM обученных на греческом языке
Модель в основе которой лежит Qwen (да-да не удивляейтесь).
ELlama1-0.7b - pretrain модель, обученная на семпле из fineweb2: fineweb2-modern-greece-sample.
import torch
from transformers import AutoModelForCausalLM, PreTrainedTokenizerFast
device = "cuda"
model_path = "dmitry315/ELlama1-0.7b"
model = AutoModelForCausalLM.from_pretrained(model_path, torch_dtype=torch.float16, trust_remote_code=True)
model.to(device)
tokenizer = PreTrainedTokenizerFast.from_pretrained(model_path, trust_remote_code=True)
text = "Γεια σας , δεν ξερω τιποτα για τον Ηροδοτο , μπορειτε να μου πειτε γι ' αυτον ;"
with torch.no_grad():
inputs = tokenizer(
text,
return_tensors="pt",
padding=True,
truncation=True,
max_length=128
).to(device)
outputs = model.generate(
inputs.input_ids,
max_length=128,
temperature=1.0,
top_p=50,
do_sample=True,
pad_token_id=tokenizer.eos_token_id,
num_return_sequences=1
)
generated_text = tokenizer.decode(
outputs[0],
skip_special_tokens=True
)
print(generated_text)
Код обучения: ELlama