- Updated dependencies to the latest including
transformersandllama-cpp-python. - Switched to
SmolLM-135M.Q8_0.ggufandgemma-2-2b-it.q8_0.ggufmodels. - Update
nertemplate to useAutoModelForCausalLMandAutoTokenizer.
transformers and llama-cpp-python.SmolLM-135M.Q8_0.gguf and gemma-2-2b-it.q8_0.gguf models.ner template to use AutoModelForCausalLM and AutoTokenizer.