Unsloth
Definition
Unsloth is a specialized tool designed to optimize the fine-tuning and deployment of …
LLM
Token maxxing
Definition
Token maxxing involves carefully crafting inputs to utilize the full capacity of a …
Self-Consistency
Definition
Primarily used with Large Language Models (LLMs), this technique improves accuracy by …
Reflection
Definition
In AI, reflection is a paradigm where a model pauses to evaluate its own generation …
Reasoning model
Definition
Unlike standard generative models focused on fluency, reasoning models prioritize …
Prompt Tuning
Definition
Prompt tuning involves adding trainable soft prompts (continuous vectors) to the input …
Pythia
Definition
Pythia is a series of open-source large language models (LLMs) created by EleutherAI, …
Phi
Definition
Phi, short for ‘Foundation models based on Teaching-Learning Paradigm’, is a …
Neural scaling law
Definition
Neural scaling laws describe the predictable power-law relationship between a …
Misinformation
Definition
Misinformation refers to false or misleading information shared without the deliberate …
Mistral
Definition
Mistral refers to a family of powerful open-weight LLMs created by the French startup …
Llama 2
Definition
Released by Meta AI in July 2023, Llama 2 represents a significant evolution in …
Llama 3
Definition
Introduced in April 2024, Llama 3 builds upon the success of Llama 2 with substantial …
Llama3.1
Definition
Released in August 2024, Llama 3.1 expands the Llama family to include a massive 405 …
Llama
Definition
Llama (Large Language Model Meta AI) is a series of foundational large language models …
Kimi K2
Definition
Kimi K2 represents a significant iteration in Moonshot AI’s series of large …
Kimi K25
Definition
Kimi K25 is an advanced iteration within the Kimi family of models produced by Moonshot …
Instruction Following
Definition
Instruction following refers to the ability of large language models and other AI systems …
Guardrails
Definition
Guardrails refer to a set of software controls and policy enforcement layers integrated …
GLM
Definition
In statistical modeling, GLM stands for Generalized Linear Models, which extend linear …
GLM MoE DSA
Definition
There is no single standard term ‘GLM MoE DSA’. However, it likely combines …
GPT-5.6
Definition
GPT-5.6 refers to a speculative or forthcoming version in the lineage of OpenAI’s …
Falcon
Definition
Falcon refers to a series of powerful large language models (LLMs) created by the …
Eagle
Definition
Eagle represents a specific architectural and engineering framework within the domain of …
DeepSeek
Definition
DeepSeek refers to a family of artificial intelligence models created by the company …
Dataset:Tiiuae/Falcon Refinedweb
Definition
RefinedWeb is a large-scale dataset of filtered web pages designed for pretraining …
Token Limit
Definition
Token limit defines the context window size constraint for large language models, …
Prompt Injection
Definition
Prompt injection exploits the way large language models interpret user instructions by …
Supervised Fine-tuning
Definition
Supervised Fine-tuning (SFT) involves taking a large pre-trained model, such as a …
Function Calling
Definition
Function calling enables large language models to interact with external tools and APIs …
Few-Shot Prompting
Definition
This method leverages the in-context learning capabilities of large language models by …
zero-shot
Definition
Zero-shot learning enables models to generalize to new categories or tasks for which no …
post-training
Definition
Post-training is a critical stage in the machine learning lifecycle that occurs after the …
Instruction Tuning
Definition
This process bridges the gap between general pre-training and specific task performance. …
Chain-of-Thought
Definition
Chain-of-Thought (CoT) prompting is a strategy where large language models are guided to …
Prompt Engineering
Definition
Prompt engineering involves crafting specific inputs, known as prompts, to elicit …