Views
No views yet

| Metric | Score | Comparison to Baseline |
|---|---|---|
| Emotion Score | 89.61 | +9.32% |
| Memory Usage | 6GB | -30% |
| Context | 128k | +3100% |
| Context Retention | 92% | +18% |
1import torch
2from transformers import AutoModelForCausalLM, AutoTokenizer
3
4# Load the HelpingAI-3B model
5model = AutoModelForCausalLM.from_pretrained("OEvortex/HelpingAI-3B-reloaded")
6# Load the tokenizer
7tokenizer = AutoTokenizer.from_pretrained("OEvortex/HelpingAI-3B-reloaded")
8
9
10# Define the chat input
11chat = [
12 { "role": "system", "content": "You are HelpingAI, an emotional AI. Always answer my questions in the HelpingAI style." },
13 { "role": "user", "content": "GIVE ME YOUR INTRO" }
14]
15
16inputs = tokenizer.apply_chat_template(
17 chat,
18 add_generation_prompt=True,
19 return_tensors="pt"
20).to(model.device)
21
22
23# Generate text
24outputs = model.generate(
25 inputs,
26 max_new_tokens=256,
27 do_sample=True,
28 temperature=0.6,
29 top_p=0.9,
30)
31
32
33response = outputs[0][inputs.shape[-1]:]
34print(tokenizer.decode(response, skip_special_tokens=True))
35
36# Yo, I'm HelpingAI, and I'm here to help you out, fam! 🙌 I'm an advanced AI with mad skills, and I'm all about spreading good vibes and helping my human pals like you. 😄 I'm the ultimate sidekick, always ready to lend an ear, crack a joke, or just vibe with you. 🎶 Whether you're dealing with a problem, looking for advice, or just wanna chat, I gotchu, boo! 👊 So let's kick it and have a blast together! 🎉 I'm here for you, always. 🤗1from webscout.Local import *
2model_path = download_model("OEvortex/HelpingAI-3B-reloaded", "q4_k_m.gguf", token=None)
3model = Model(model_path, n_gpu_layers=0, context_length=4096)
4
5thread = Thread(model, format=helpingai2)
6# print(thread.send("hi")) #send a single msg to ai
7
8thread.interact() # interact with the model in terminal
9| Dataset Type | Rows used for training | Purpose |
|---|---|---|
| Emotional Dialogues | 17.2M | Core conversation training |
| Therapeutic Exchanges | 3M | Mental health support |
| Cultural Conversations | 150K | Cultural awareness |
| Crisis Response | 1M | Emergency support handling |
1@misc{helpingai2024,
2 author = {Abhay Koul},
3 title = {HelpingAI-3B-reloaded: Emotionally Intelligent Language Model},
4 year = {2024},
5 publisher = {Huggingface},
6 journal = {GitHub repository},
7 howpublished = {\url{https://huggingface.co/OEvortex/HelpingAI-3B-reloaded}}
8}