S claudeskill.wiki
ai-research

post-training-simpo

تحسين التفضيلات البسيط لمحاذاة نماذج اللغة الكبيرة (LLM). بديل بدون مرجع لـ DPO بأداء أفضل (+6.4 نقاط على AlpacaEval 2.0). لا حاجة إلى نموذج مرجعي، أكثر كفاءة من DPO. يُستخدم لمحاذاة التفضيلات عندما ترغب في تدريب أبسط وأسرع من DPO/PPO.

١٠

SimPO - Simple Preference Optimization

Quick start

SimPO is a reference-free preference optimization method that outperforms DPO without needing a reference model.

Installation:

# Create environment
conda create -n simpo python=3.10 && conda activate simpo

# Install PyTorch 2.2.2
# Visit: https://pytorch.org/get-started/locally/

# Install alignment-handbook
git clone https://github.com/huggingface/alignment-handbook.git
cd alignment-handbook
python -m pip install .

# Install Flash Attention 2
python -m pip install flash-attn --no-build-isolation

Training (Mistral 7B):

ACCELERATE_LOG_LEVEL=info accelerate launch \
  --config_file accelerate_configs/deepspeed_zero3.yaml \
  scripts/run_simpo.py \
  training_configs/mistral-7b-base-simpo.yaml

Common workflows

Workflow 1: Train from base model (Mistral 7B)

Config (mistral-7b-base-simpo.yaml):

# Model
model_name_or_path: mistralai/Mistral-7B-v0.1
torch_dtype: bfloat16

# Dataset
dataset_mixer:
  HuggingFaceH4/ultrafeedback_binarized: 1.0
dataset_splits:
  - train_prefs
  - test_prefs

# SimPO hyperparameters
beta: 2.0                  # Reward scaling (2.0-10.0)
gamma_beta_ratio: 0.5       # Target margin (0-1)
loss_type: sigmoid          # sigmoid or hinge
sft_weight: 0.0             # Optional SFT regularization

# Training
learning_rate: 5e-7         # Critical: 3e-7 to 1e-6
num_train_epochs: 1
per_device_train_batch_size: 1
gradient_accumulation_steps: 8

# Output
output_dir: ./outputs/mistral-7b-simpo

Launch training:

accelerate launch --config_file accelerate_configs/deepspeed_zero3.yaml \
  scripts/run_simpo.py training_configs/mistral-7b-base-simpo.yaml

Workflow 2: Fine-tune instruct model (Llama 3 8B)

Config (llama3-8b-instruct-simpo.yaml):

model_name_or_path: meta-llama/Meta-Llama-3-8B-Instruct

dataset_mixer:
  argilla/ultrafeedback-binarized-preferences-cleaned: 1.0

beta: 2.5
gamma_beta_ratio: 0.5
learning_rate: 5e-7
sft_weight: 0.1             # Add SFT loss to preserve capabilities

num_train_epochs: 1
per_device_train_batch_size: 2
gradient_accumulation_steps: 4
output_dir: ./outputs/llama3-8b-simpo

Launch:

accelerate launch --config_file accelerate_configs/deepspeed_zero3.yaml \
  scripts/run_simpo.py training_configs/llama3-8b-instruct-simpo.yaml

Workflow 3: Reasoning-intensive tasks (lower LR)

For math/code tasks:

model_name_or_path: deepseek-ai/deepseek-math-7b-base

dataset_mixer:
  argilla/distilabel-math-preference-dpo: 1.0

beta: 5.0                   # Higher for stronger signal
gamma_beta_ratio: 0.7       # Larger margin
learning_rate: 3e-7         # Lower LR for reasoning
sft_weight: 0.0

num_train_epochs: 1
per_device_train_batch_size: 1
gradient_accumulation_steps: 16

When to use vs alternatives

Use SimPO when:

  • Want simpler training than DPO (no reference model)
  • Have preference data (chosen/rejected pairs)
  • Need better performance than DPO
  • Limited compute resources
  • Single-node training sufficient

Algorithm selection:

  • SimPO: Simplest, best performance, no reference model
  • DPO: Need reference model baseline, more conservative
  • PPO: Maximum control, need reward model, complex setup
  • GRPO: Memory-efficient RL, no critic

Use alternatives instead:

  • OpenRLHF: Multi-node distributed training, PPO/GRPO
  • TRL: Need multiple methods in one framework
  • DPO: Established baseline comparison

Common issues

Issue: Loss divergence

Reduce learning rate:

learning_rate: 3e-7  # Reduce from 5e-7

Reduce beta:

beta: 1.0  # Reduce from 2.0

Issue: Model forgets capabilities

Add SFT regularization:

sft_weight: 0.1  # Add SFT loss component

Issue: Poor preference separation

Increase beta and margin:

beta: 5.0            # Increase from 2.0
gamma_beta_ratio: 0.8  # Increase from 0.5

Issue: OOM during training

Reduce batch size:

per_device_train_batch_size: 1
gradient_accumulation_steps: 16  # Maintain effective batch

Enable gradient checkpointing:

gradient_checkpointing: true

Advanced topics

Loss functions: See references/loss-functions.md for sigmoid vs hinge loss, mathematical formulations, and when to use each.

Hyperparameter tuning: See references/hyperparameters.md for beta, gamma, learning rate selection guide, and model-size-specific recommendations.

Dataset preparation: See references/datasets.md for preference data formats, quality filtering, and custom dataset creation.

Hardware requirements

  • GPU: NVIDIA A100/H100 recommended
  • VRAM:
    • 7B model: 1× A100 40GB (DeepSpeed ZeRO-3)
    • 8B model: 2× A100 40GB
    • 70B model: 8× A100 80GB
  • Single-node: DeepSpeed ZeRO-3 sufficient
  • Mixed precision: BF16 recommended

Memory optimization:

  • DeepSpeed ZeRO-3 (default config)
  • Gradient checkpointing
  • Flash Attention 2

Resources

التثبيت

npx claude-code-templates@latest --skill ai-research/post-training-simpo

Quick start

  1. Install Claude Code if you have not already.
  2. Copy the Install command from this page and run it in your project directory.
  3. In Claude Code, load or mention the skill when your task matches what the skill is for.

Documentation

Use the links below for agent skills, troubleshooting, and official examples.

موارد