Results for “efficient training”

Train language models with SimPO using pairs of preferred and rejected answers. Configure preference training without a separate reference model, adapt settings for base or instruction models, and address unstable training, memory limits, or lost capabilities.

• Training a base model with preferred and rejected answers
• Further tuning an instruction model with preference data
• Adjusting training settings for math and code tasks

Guide language model training with SimPO using pairs of preferred and rejected answers. Covers configuration, learning rate tuning, and memory troubleshooting for preference alignment, which adjusts model responses toward human preferences without a separate reference model.

• Training on preferred and rejected answer pairs
• Exploring an alternative to DPO without a reference model
• Tuning preference training for math and code tasks

Helps write and review PyTorch models, training loops, and data loading code. Covers repeatable experiments, saving training progress, and techniques for reducing GPU memory use or improving training speed, with practical code examples.

• Writing or reviewing PyTorch training code
• Investigating inconsistent experiment results
• Improving data loading and GPU memory use

Get guidance on adapting LLMs with Unsloth, including implementation, API usage, and debugging. Covers LoRA and QLoRA, methods for adjusting models with fewer trainable parameters, with a focus on training speed and memory use.

• Learning how to fine-tune models with Unsloth
• Exploring LoRA and QLoRA for efficient training
• Implementing features using Unsloth APIs

Train Mixture of Experts models, which activate selected specialist networks for each input. The skill guides architecture, input routing, workload balancing, and GPU distribution using DeepSpeed or HuggingFace, with examples for training configuration and inference optimization.

• Designing a model that activates only selected expert networks
• Balancing workloads across expert networks
• Distributing experts across multiple GPUs

Get guidance on fine-tuning language models with Unsloth, including LoRA and QLoRA methods for training with fewer resources. The skill points to documentation for understanding APIs, implementing training code, debugging problems, and exploring speed and memory optimizations.

• Exploring faster training with lower memory use
• Learning about LoRA and QLoRA fine-tuning
• Checking Unsloth features and APIs

Simple Preference Optimization for LLM alignment. Reference-free alternative to DPO with better performance (+6.4 points on AlpacaEval 2.0). No reference model needed, more efficient than DPO. Use for preference alignment when want simpler, faster training than DPO/PPO.

The English summary is being prepared. Showing the original description.

Build and train Mixture of Experts models, which activate selected specialist networks for each input. Use DeepSpeed or HuggingFace workflows to configure routing, balance workloads, distribute experts across GPUs, and optimize inference.

• Designing MoE models for a limited compute budget
• Implementing Mixtral-style expert routing
• Adjusting expert workloads and training stability

Reduce GPU memory use by loading language models in 8-bit or 4-bit formats with bitsandbytes and HuggingFace Transformers. Includes configuration examples, QLoRA fine-tuning of compressed models, and optimizers that reduce memory needed during training.

• Load larger models within limited GPU memory
• Compare 8-bit and 4-bit model settings
• Fine-tune a model with QLoRA

Fine-tune large language models with LoRA and QLoRA while limiting GPU memory use. Guides training settings and adapter management, using small sets of task-specific weights that can be saved, switched, or merged into a base model.

• Fine-tuning an LLM with limited GPU memory
• Choosing LoRA settings and training layers
• Switching task-specific adapters on one base model

Helps reduce GPU memory use by loading large language models in 4-bit or 8-bit formats, a technique called quantization. Covers bitsandbytes configuration, memory checks, QLoRA fine-tuning, and training with memory-saving optimizers.

• Load a larger model within available GPU memory
• Compare 4-bit and 8-bit quantization settings
• Fine-tune a quantized model with QLoRA

Compress large language models to 4-bit weights to reduce GPU memory needs and accelerate text generation. Covers loading compressed models, converting your own models, choosing settings, and preparing memory-efficient fine-tuning with transformers and PEFT.

• Run language models with limited GPU memory
• Convert your own model to 4-bit weights
• Compare compression settings for quality and speed

Train GPT-2 scale models (~124M parameters) efficiently on a single GPU. Covers GPT-124M architecture, tokenized dataset loading (e.g., HuggingFace Hub shards), modern optimizers (Muon, AdamW), mixed precision training, and training loop implementation.

The English summary is being prepared. Showing the original description.

Simple Preference Optimization para alinhamento de LLMs. Alternativa sem modelo de referência ao DPO com melhor desempenho (+6.4 pontos no AlpacaEval 2.0). Sem modelo de referência necessário, mais eficiente que DPO. Use para alinhamento de preferências quando quer treinamento mais simples e rápido que DPO/PPO.

The English summary is being prepared. Showing the original description.

Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace. Use when training large-scale models with limited compute (5× cost reduction vs dense models), implementing sparse architectures like Mixtral 8x7B or DeepSeek-V3, or scaling model capacity without proportional compute increase. Covers MoE architectures, routing mechanisms, load balancing, expert parallelism, and inference optimization.

The English summary is being prepared. Showing the original description.

Reduce the memory needed to run large language models with GPTQ quantization, which stores model weights at lower precision. Covers loading compressed models, quantizing your own models, choosing execution backends, and connecting to memory-efficient fine-tuning.

• Run a language model within limited GPU memory
• Load a pre-quantized model for text generation
• Quantize and save your own trained model

Extend the amount of text a language model can process at once. Compares RoPE, YaRN, ALiBi, and position interpolation, with implementation examples, additional training guidance, and memory optimization approaches for long document processing.

• Extend an existing model to handle longer inputs
• Compare methods for representing token positions
• Plan additional training with long documents

Helps fine-tune large language models with LoRA and QLoRA, methods that train small sets of additional parameters. Covers memory-conscious training, configuration choices, saving adapters, and switching task-specific adapters on a shared base model.

• Fine-tune a model with limited GPU memory
• Choose LoRA settings for a specific task
• Switch task-specific adapters on one base model

Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization

The English summary is being prepared. Showing the original description.

Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization

The English summary is being prepared. Showing the original description.

Treinar modelos de Mixture of Experts (MoE) usando DeepSpeed ou HuggingFace. Use ao treinar modelos em larga escala com computação limitada (redução de 5× em custos vs modelos densos), implementar arquiteturas esparsas como Mixtral 8x7B ou DeepSeek-V3, ou escalar capacidade de modelo sem aumento proporcional de computação. Cobre arquiteturas MoE, mecanismos de roteamento, balanceamento de carga, paralelismo de especialistas e otimização de inferência.

The English summary is being prepared. Showing the original description.

Simple Preference Optimization for LLM alignment. Reference-free alternative to DPO with better performance (+6.4 points on AlpacaEval 2.0). No reference model needed, more efficient than DPO. Use for preference alignment when want simpler, faster training than DPO/PPO.

The English summary is being prepared. Showing the original description.

This skill should be used at the start of any computationally intensive scientific task to detect and report available system resources (CPU cores, GPUs, memory, disk space). It creates a JSON file with resource information and strategic recommendations that inform computational approach decisions such as whether to use parallel processing (joblib, multiprocessing), out-of-core computing (Dask, Zarr), GPU acceleration (PyTorch, JAX), or memory-efficient strategies. Use this skill before running analyses, training models, processing large datasets, or any task where resource constraints matter.

The English summary is being prepared. Showing the original description.

This skill should be used at the start of any computationally intensive scientific task to detect and report available system resources (CPU cores, GPUs, memory, disk space). It creates a JSON file with resource information and strategic recommendations that inform computational approach decisions such as whether to use parallel processing (joblib, multiprocessing), out-of-core computing (Dask, Zarr), GPU acceleration (PyTorch, JAX), or memory-efficient strategies. Use this skill before running analyses, training models, processing large datasets, or any task where resource constraints matter.

The English summary is being prepared. Showing the original description.