How to use mcmonkey/google_t5-v1_1-xxl_encoderonly with Transformers:
# Load model directly from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("mcmonkey/google_t5-v1_1-xxl_encoderonly") model = AutoModel.from_pretrained("mcmonkey/google_t5-v1_1-xxl_encoderonly")
when I use the config.json here load model.safetensors, but report the error_IncompatibleKeys(missing_keys=['encoder.embed_tokens.weight'], unexpected_keys=[])
Is it that safetensor does not save the encoder's embedding weights?
· Sign up or log in to comment
