Инструктивная модель на основе Vikhr-Qwen-2.5-1.5B-Instruct, обученная на русскоязычном сгенерированном датасете. Создана для ответа на запросы в роли волшебного психолога из Хоггвартса.

Пример использования:

def generate_answer(model, tokenizer, prompt):

  inputs = tokenizer.apply_chat_template([{"role": "system", "content": BASE_SYSTEM_PROMPT},{"role": "user", "content": prompt}],
                                       add_generation_prompt=True,
                                       tokenize=True,
                                       return_tensors="pt",
                                       return_dict=True
                                       ).to('cuda')


  gen_kwargs = {"temperature":0.5,
    "top_p":0.8,
    "max_length":512,
    "repetition_penalty":1.05,
     "do_sample": True}


  with torch.no_grad():
      outputs = model.generate(**inputs, **gen_kwargs)
      outputs = outputs[:, inputs['input_ids'].shape[1]:]
      return tokenizer.decode(outputs[0], skip_special_tokens=True)
Downloads last month
230
Safetensors
Model size
1.54B params
Tensor type
F32
·
Inference Providers NEW
This model is not currently available via any of the supported third-party Inference Providers, and the model is not deployed on the HF Inference API.

Model tree for divisee/Magic-psy-LoRA-Qwen-Finetuned

Quantizations
2 models