runtime error

Exit code: 1. Reason: from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing BertForMaskedLM from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Inicializando Vortex Betina... Tentando baixar pesos do repositório: reynaldo22/betina-perfect-2025... ⚠️ Falha ao baixar do Hugging Face: 401 Client Error. (Request ID: Root=1-69248e54-157a35561452634764bb38ee;8557bda8-b879-42d2-9c82-6a19dd97c450) Cannot access gated repo for url https://huggingface.co/reynaldo22/betina-perfect-2025/resolve/main/vortex.pt. Access to model reynaldo22/betina-perfect-2025 is restricted. You must have access to it and be authenticated to access it. Please log in. Tentando carregar localmente... Carregando pesos locais de .... Erro ao carregar pesos de .: Error(s) in loading state_dict for VortexBetinaAntiHalluc: Missing key(s) in state_dict: "projector.1.bias". Unexpected key(s) in state_dict: "sigma", "rho", "beta", "gamma", "noise_scale", "projector2.weight", "adversary.weight". ⚠️ AVISO: Pesos treinados não encontrados. Usando inicialização aleatória. O modelo vai rodar, mas as respostas da Betina serão aleatórias até você treinar. Traceback (most recent call last): File "/app/app.py", line 212, in <module> with gr.Blocks(title="Betina 2.0 - Anti-Hallucination AI", css=custom_css) as demo: File "/usr/local/lib/python3.10/site-packages/gradio/blocks.py", line 1071, in __init__ super().__init__(render=False, **kwargs) TypeError: BlockContext.__init__() got an unexpected keyword argument 'css' model.safetensors: 26%|██▌ | 113M/438M [00:01<00:02, 112MB/s] model.safetensors: 100%|██████████| 438M/438M [00:02<00:00, 199MB/s] model.safetensors: 100%|██████████| 438M/438M [00:02<00:00, 188MB/s]

Container logs: