nanodex-100k-1m

A 100,064-parameter decoder-only language model pre-trained from scratch on fineweb-edu, using the HyperDex Trainer Space.

Architecture

A standard LlamaForCausalLM decoder-only transformer — SiLU MLP, RMSNorm, rotary position embeddings, grouped-query attention, tied embeddings, no biases — scaled down in width and depth to fit the parameter budget.

Parameters	100,064
Hidden size	32
Layers	2
Attention heads	2 (KV: 1)
FFN size	147
Context length	512
Vocab	2,048 (custom BPE trained on fineweb-edu)

Training

Tokens seen	983,040
Steps	15
Tokens / step	65,536
Optimizer	AdamW(0.9, 0.95) wd=0.1 clip=1.0
LR schedule	warmup 2% + cosine to 10% (peak 5e-03)
Final loss	6.8223 (ppl 918.1)
Wall time	0.0 min
Trained by	@GGUFGuy

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

tok = AutoTokenizer.from_pretrained("GGUFGuy/nanodex-100k-1m")
model = AutoModelForCausalLM.from_pretrained("GGUFGuy/nanodex-100k-1m")

ids = tok("The mitochondria is", return_tensors="pt").input_ids
print(tok.decode(model.generate(ids, max_new_tokens=60, do_sample=True,
                                temperature=0.8, top_k=50)[0]))

Caveats

This is a small-scale research artifact. At this parameter count and token budget the model learns word shapes, common collocations and a little syntax — it is not a useful assistant and its output is not factual. It exists to make "pre-train a transformer from scratch" something you can actually watch happen.