Results for “l4t”

Manages suspected mild central hypothyroidism in patients with pituitary disease and low-normal free thyroxine (fT4). Initiates levothyroxine (L-T4) when suggestive symptoms are present or when serial fT4 shows a decrease of 20% or more.

Evaluates central hypothyroidism (CH) patients for adrenal insufficiency (AI) prior to initiating levothyroxine (L‑T4) replacement. Triggers include consideration of L‑T4 start in central hypothyroidism needing thyroid replacement.

Monitors euthyroid adults with growth hormone deficiency (GHD) who are initiating GH replacement therapy for the development of central hypothyroidism (CH). Initiates levothyroxine (L-T4) if free thyroxine (fT4) falls below the laboratory reference range; triggers include GHD patient starting GH replacement.

Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models (7B-70B) on limited GPU memory, when you need faster inference than GPTQ with better accuracy preservation, or for instruction-tuned and multimodal models. MLSys 2024 Best Paper Award winner.

Post-training 4-bit quantization for LLMs with minimal accuracy loss. Use for deploying large models (70B, 405B) on consumer GPUs, when you need 4× memory reduction with <2% perplexity degradation, or for faster inference (3-4× speedup) vs FP16. Integrates with transformers and PEFT for QLoRA fine-tuning.

Post-training 4-bit quantization for LLMs with minimal accuracy loss. Use for deploying large models (70B, 405B) on consumer GPUs, when you need 4× memory reduction with <2% perplexity degradation, or for faster inference (3-4× speedup) vs FP16. Integrates with transformers and PEFT for QLoRA fine-tuning.

Determines levothyroxine dosing for central hypothyroidism to achieve serum free T4 in the mid to upper half of the reference range. Triggers include confirmed central hypothyroidism requiring thyroid hormone replacement when initiating L‑4 therapy.

Flash a promoted BSP image to a Jetson device in RCM mode using NVIDIA's flash.sh or l4t_initrd_flash.sh toolchain.

Monitor free T4 and adjust levothyroxine after starting antiepileptic drugs in central hypothyroidism

Quantize large language models to 4-bit with minimal accuracy loss using GPTQ, enabling deployment of 70B+ models on consumer GPUs with 4× memory reduction and 3-4× faster inference.

Provides patterns for hydration safety, performance, route rules, lazy loading, and SSR-safe data fetching in Nuxt 4 apps.

GPT-4V(ision) System Card

Nuxt 4 app patterns for hydration safety, performance, route rules, lazy loading, and SSR-safe data fetching with useFetch and useAsyncData.

Quantize large language models to 4-bit precision using activation-aware weight quantization, reducing memory footprint and speeding up inference with minimal accuracy loss.

This skill guides perioperative levothyroxine management for adults with preoperative central hypothyroidism undergoing non-emergency surgery. Trigger phrases include "preoperative central hypothyroidism patient requiring non-emergency surgery" and "use when managing preoperative CH patient".

Nuxt 4 app patterns for hydration safety, performance, route rules, lazy loading, and SSR-safe data fetching with useFetch and useAsyncData.

Automate multi-turn adversarial conversations against LLM agents using Microsoft PyRIT, including Crescendo and Tree-of-Attacks-with-Pruning (TAP) attack chains with scorer feedback loops.

Quantize LLMs to 8-bit or 4-bit for 50-75% memory reduction with minimal accuracy loss using bitsandbytes. Supports INT8, NF4, FP4 formats, QLoRA training, and 8-bit optimizers.

Liquid Network: federated sidechain by Blockstream. Confidential Transactions, asset issuance (LBTC, USDt-Liquid, others), 2-min blocks, n-of-m federation peg-out. Elements codebase. USE WHEN: building on Liquid, peg-in/peg-out integrations, designing CT-based privacy.

Selects the optimal LLM model and provider for each task based on complexity, cost budget, and capability requirements. Routes cheap tasks to Haiku/GPT-4o-mini and complex tasks to Sonnet/Opus/o1. Use when deciding which model to call, optimizing LLM costs, or building multi-model agent systems. Activate on "which model", "model selection", "route to model", "LLM cost", "model routing", "cheap vs expensive model". NOT for prompt engineering (use prompt-engineer), model fine-tuning, or training custom models.

Meta's 7-8B specialized moderation model for LLM input/output filtering. 6 safety categories - violence/hate, sexual content, weapons, substances, self-harm, criminal planning. 94-95% accuracy. Deploy with vLLM, HuggingFace, Sagemaker. Integrates with NeMo Guardrails.

SLF4J - Simple Logging Facade for Java. Standard logging API that abstracts underlying implementation (Logback, Log4j2). Provides parameterized logging and MDC support. USE WHEN: user mentions "slf4j", "java logging api", "parameterized logging", asks about "how to log in Java", "logger facade", "MDC in java", "logging best practices java" DO NOT USE FOR: Logback configuration - use `logback` instead, Log4j2 configuration - use Log4j2 skill, Node.js logging - use `winston` or `pino` instead, Python logging - use `python-logging` instead

Accelerate LLM inference using speculative decoding, Medusa multiple heads, and lookahead decoding techniques. Use when optimizing inference speed (1.5-3.6× speedup), reducing latency for real-time applications, or deploying models with limited compute. Covers draft models, tree-based attention, Jacobi iteration, parallel token generation, and production deployment strategies.

Expert en IoT avancé (MQTT, edge, device management, digital twin, security, DZ infrastructure)

Optimizes LLM inference with NVIDIA TensorRT for maximum throughput and lowest latency. Use for production deployment on NVIDIA GPUs (A100/H100), when you need 10-100x faster inference than PyTorch, or for serving models with quantization (FP8/INT4), in-flight batching, and multi-GPU scaling.

Fine-tunes and evaluates OpenVLA-OFT and OpenVLA-OFT+ policies for robot action generation with continuous action heads, LoRA adaptation, and FiLM conditioning on LIBERO simulation and ALOHA real-world setups. Use when reproducing OpenVLA-OFT paper results, training custom VLA action heads (L1 or diffusion), deploying server-client inference for ALOHA, or debugging normalization, LoRA merge, and cross-GPU issues.

Meta's 86M prompt injection and jailbreak detector. Filters malicious prompts and third-party data for LLM apps. 99%+ TPR, <1% FPR. Fast (<2ms GPU). Multilingual (8 languages). Deploy with HuggingFace or batch processing for RAG security.

SVIT: Scaling up Visual Instruction Tuning

Unified LLM Gateway - One API for 70+ AI models. Route to GPT, Claude, Gemini, Qwen, Deepseek, Grok and more with a single API key. Use when: the user needs model routing, provider setup, or Chinese LLM access guidance.

Accelerate LLM inference using speculative decoding, Medusa multiple heads, and lookahead decoding techniques. Use when optimizing inference speed (1.5-3.6× speedup), reducing latency for real-time applications, or deploying models with limited compute. Covers draft models, tree-based attention, Jacobi iteration, parallel token generation, and production deployment strategies.

Quantize large language models to 4-bit using activation-aware weight quantization, achieving ~3x speedup with minimal accuracy loss for deployment on limited GPU memory.

Train large-scale MoE models with FP8/INT4 low-precision RL, speculative decoding, and train-inference alignment using the miles framework.

Review L2 session facts and promote valuable ones to L1 Atomic Memory. Use at end of session or when L2 accumulates facts worth keeping permanently. Runs add-fact.sh for each promoted fact.

Fast-executing ML engineer for experiments. Use when the user asks to talk to Jett, requests the ML engineer, or needs experiment implementation.