nanowhale

A small ~110M parameter language model implementing the DeepSeek-V4 architecture, fine-tuned for chat/instruction following. Trained from scratch — no weights from DeepSeek-V4 were used.

• Pretrained base model: HuggingFaceTB/nanowhale-100m-base
• This model: SFT on HuggingFaceTB/smol-smoltalk
• Training code: github.com/huggingface/nanowhale

Architecture

This model implements key DeepSeek-V4 innovations at a miniature scale:

Training

Stage 1: Pretraining

• Dataset: HuggingFaceFW/fineweb-edu
• Steps: 5,000 | Tokens: ~2.6B
• Batch: 32 effective (8 × 4 GA) | Seq length: 2,048
• LR: 6e-4, cosine, 3% warmup
• Precision: bf16 mixed

Stage 2: SFT (this model)

• Dataset: HuggingFaceTB/smol-smoltalk (460K conversations)
• Steps: 3,000 | Tokens: ~72.7M
• Batch: 32 effective (8 × 4 GA) | Seq length: 2,048
• LR: 2e-5, cosine, 5% warmup
• Precision: fp32

Metrics

Usage

import torch
from safetensors.torch import load_file
from transformers import AutoConfig, AutoModelForCausalLM, AutoTokenizer
from huggingface_hub import hf_hub_download

# Load model (recommended: manual load for reliability)
config = AutoConfig.from_pretrained("HuggingFaceTB/nanowhale-100m", trust_remote_code=True)
model = AutoModelForCausalLM.from_config(config, trust_remote_code=True).float()

# Download and load weights
weights_path = hf_hub_download("HuggingFaceTB/nanowhale-100m", "model.safetensors")
state_dict = load_file(weights_path)
model.load_state_dict(state_dict, strict=True)
model = model.cuda().eval()

tokenizer = AutoTokenizer.from_pretrained("HuggingFaceTB/nanowhale-100m")

# Chat
messages = [{"role": "user", "content": "What are 3 benefits of exercise?"}]
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
input_ids = tokenizer.encode(prompt, return_tensors="pt").cuda()
output = model.generate(input_ids, max_new_tokens=200, temperature=0.7, top_p=0.9,
                        pad_token_id=tokenizer.eos_token_id)
print(tokenizer.decode(output[0][input_ids.shape[1]:], skip_special_tokens=True))

Limitations

• Tiny model: 110M params with 129K vocabulary — most capacity goes to embeddings. Generations are often incoherent or factually wrong.
• Undertrained: Only 5K pretrain + 3K SFT steps. Production models train for 100K+ steps on trillions of tokens.
• Educational purpose: This model demonstrates the DeepSeek-V4 architecture at small scale. It is not suitable for any production use.
• bf16 NaN: Use fp32 — the Hyper-Connections architecture produces values that overflow bf16 range at this scale.
• Custom code: Requires trust_remote_code=True.

Hardware

Trained on 1× NVIDIA H100 80GB.

License

Apache-2.0