You can deploy this model and start earning money today!

AtlasFoundation/
randygpt-s

Name: randygpt-s
Availability: InStock

Views

No views yet

randyGPT — model-s

A GPT-style language model trained from scratch in Rust on Project Gutenberg.

Model Details


Architecture	Transformer (causal LM)
Parameters	1.99M
Layers	8
Heads	4
Embedding dim	128
Context window	256 tokens
Vocab size	1500 (BPE)
Training iters	1800
Best val loss	4.4794

Training

Trained on ~103MB of cleaned Project Gutenberg text (114 public domain books) with BPE-1500 tokenization, AdamW optimizer, cosine LR decay, and ReduceLROnPlateau. Metal GPU via Candle on Apple Silicon.

Usage

python

from modeling_randygpt import RandyGPTConfig, RandyGPTForCausalLM
from tokenizer_randygpt import RandyGPTTokenizer
from safetensors.torch import load_file
import torch

# Load
cfg   = RandyGPTConfig.from_pretrained("MonumentalSystems/randygpt-s")
model = RandyGPTForCausalLM(cfg)
state = load_file("model.safetensors")
model.load_state_dict(state, strict=True)
model.eval()

tok = RandyGPTTokenizer.from_file("tokenizer.json")

# Generate
prompt  = "Once upon a time"
ids     = torch.tensor([tok.encode(prompt)], dtype=torch.long)
out_ids = model.generate_text(ids, max_new_tokens=200, temperature=0.8)
print(tok.decode(out_ids[0].tolist()))

Source

Trained with randyGPT — a GPT implementation in Rust with Metal GPU acceleration.

AtlasFoundation/randygpt-s