gguf: https://huggingface.co/mondk/Msh-Tiny-47M-GGUF
msh-tiny
A tiny (~47M parameter) GPT-2-architecture chat model, trained completely from scratch — no pretrained base model. Custom BPE tokenizer trained from zero, custom transformer trained from random initialization, then converted into a standard GPT2LMHeadModel for compatibility with the wider ecosystem.
Limitations
Trained from random initialization on a modest amount of data with limited compute — a small educational project, not a production-quality assistant. Expect reliable chat formatting but limited/inconsistent knowledge and occasional incoherent answers.
Prompt format
<|user|>
{your message}
<|assistant|>
The model stops generating at <|end|>.
Usage
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
model = AutoModelForCausalLM.from_pretrained("mondk/Msh-Tiny-47M")
tokenizer = AutoTokenizer.from_pretrained("mondk/Msh-Tiny-47M")
prompt = "<|user|>\nhi\n<|assistant|>\n"
input_ids = tokenizer(prompt, return_tensors="pt").input_ids
output = model.generate(input_ids, max_new_tokens=150, do_sample=True, temperature=0.7, top_k=40)
print(tokenizer.decode(output[0][input_ids.shape[1]:], skip_special_tokens=True))
Training data
Combining 15 well-known open instruction/chat datasets plus a small hand-written set of everyday chit-chat (greetings, thanks, small talk):
- mondk/Greetings-hi-for-train-Msh-v2
- tatsu-lab/alpaca
- databricks/databricks-dolly-15k
- teknium/OpenHermes-2.5
- m-a-p/Code-Feedback
- OpenAssistant/oasst1
- FreedomIntelligence/medical-o1-reasoning-SFT
- glaiveai/glaive-function-calling-v2
- openai/openai_humaneval
- HuggingFaceH4/no_robots
- open-thoughts/OpenThoughts-114k
- HuggingFaceH4/ultrachat_200k
- google-research-datasets/poem_sentiment
- CohereLabs/aya_dataset
- sentence-transformers/natural-questions