transformer-1b-chat

Transformer-1B-Chat

A 1.1 billion parameter decoder-only language model trained entirely from scratch -- pretraining, supervised fine-tuning, and preference alignment -- on 8x NVIDIA H100 GPUs.

Model Details

Architecture Highlights

• RoPE (Rotary Position Embeddings) with theta=10,000
• Grouped Query Attention (GQA) -- 4:1 query-to-KV head ratio for efficient inference
• SwiGLU Feed-Forward Network
• RMSNorm in a pre-norm configuration
• Flash Attention 2 via PyTorch SDPA

Training Pipeline

This model was built through a complete 3-stage training pipeline:

Stage 1: Pretraining

Stage 2: Supervised Fine-Tuning (SFT)

Stage 3: Direct Preference Optimization (DPO)

Hardware

• 8x NVIDIA H100 80GB HBM3
• Distributed Strategy: PyTorch DDP (DistributedDataParallel)
• Communication: NCCL
• Mixed Precision: BF16 autocast
• Total Training Time: ~13.5 hours (all 3 stages)

Chat Template

The model uses a simple chat template with special tokens:

<|user|>
Your message here
<|end|>
<|assistant|>
Model response here
<|end|>

Special Tokens

Limitations

• 1.1B parameters -- smaller models have inherent limitations in reasoning depth and factual accuracy
• Trained on English data only
• May generate plausible-sounding but incorrect information
• The DPO alignment is single-epoch; additional iterations could improve quality
• Not safety-tuned beyond what the UltraFeedback dataset provides

Training Code

The full training code is open-sourced alongside this model.

model/
  config.py          # Model and training hyperparameters
  transformer.py     # Full transformer implementation from scratch
  data.py            # Pretraining data pipeline (FineWeb-Edu)
  sft_data.py        # SFT data pipeline (UltraChat)
  dpo_data.py        # DPO data pipeline (UltraFeedback)
train.py             # Pretraining script (DDP, 8-GPU)
train_sft.py         # SFT script
train_dpo.py         # DPO script
chat.py              # Interactive chat interface
export_to_hf.py      # Export to HuggingFace format

License

Apache 2.0