EMO-1.5B:
EMO-1.5B is a powerful language model designed to engage in emotionally intelligent conversations.
Overview
EMO-1.5B is a state-of-the-art conversational AI model with 1.5 billion parameters. It has been fine-tuned on a diverse corpus of emotional narratives, enabling it to perceive and respond to the emotional undertones present in user inputs. Whether you're seeking comfort, motivation, or simply an empathetic listener, EMO-1.5B is here to provide emotional support and guidance.
Key Features
- Emotional Intelligence: EMO-1.5B can recognize and respond to various emotions, such as sadness, joy, anger, and fear, with appropriate emotional responses.
- Contextual Understanding: The model considers the broader context of the conversation to provide relevant and emotionally resonant responses.
- Empathetic Dialogue: EMO-1.5B excels at active listening, validating emotions, and offering compassionate advice or consolation when needed.
- Adaptive Persona: The model can adapt its persona and communication style to match the user's emotional state, providing a personalized and tailored experience.
Usage
You can easily interact with EMO-1.5B using the provided example code:
from transformers import AutoModelForCausalLM, AutoTokenizer
device = "cuda" # the device to load the model onto
model = AutoModelForCausalLM.from_pretrained(
"OEvortex/EMO-1.5B",
torch_dtype="auto",
device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained("OEvortex/EMO-1.5B")
prompt = "Imagine you're helping someone who is feeling overwhelmed. How do you feel in this situation?"
messages = [
{"role": "system", "content": "You are a helpful and emotional assistant that will always respond in EMO style"},
{"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(device)
generated_ids = model.generate(
model_inputs.input_ids,
max_new_tokens=512
)
generated_ids = [
output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
print(response)