Unsloth
Definition
Unsloth is a specialized tool designed to optimize the fine-tuning and deployment of …
Training
Reciprocal human machine learning
Definition
This approach moves beyond simple human-in-the-loop labeling. It involves bidirectional …
Offline learning
Definition
Also known as batch learning, offline learning involves training machine learning models …
Multi-task Learning
Definition
This technique leverages the inductive bias shared among related tasks to enhance …
Mixed Precision Training
Definition
Mixed Precision Training (MPT) combines half-precision (FP16) and full-precision (FP32) …
Learning curve
Definition
Typically, a learning curve displays training and validation scores on the y-axis against …
Knowledge Distillation
Definition
Knowledge distillation is a machine learning method used to compress a large, complex …
Imatrix
Definition
Imatrix, short for Importance Matrix, is a technique primarily associated with GGML-based …
Hyperparameter
Definition
Unlike model parameters (weights and biases) that are learned from data during training, …
Hyperparameter Tuning
Definition
Hyperparameter tuning involves evaluating different sets of hyperparameters to find the …
Grokking
Definition
Grokking refers to a counter-intuitive behavior observed in deep learning where a model …
Finetuned
Definition
Finetuning refers to the technique of taking a model that has already been trained on a …
Epoch
Definition
In machine learning, an epoch represents a single iteration over the entire training …
Eager learning
Definition
In eager learning, the system constructs a general target function or model based on the …
Early Stopping
Definition
Early stopping is a form of regularization used primarily in iterative training processes …
Domain Adaptation
Definition
Domain adaptation addresses the challenge when training and testing data come from …
Constitutional AI
Definition
Constitutional AI is a framework for aligning large language models with human values …
Apprenticeship learning
Definition
Apprenticeship learning, also known as inverse reinforcement learning from …
Adversarial machine learning
Definition
This field encompasses both offensive techniques to break models and defensive strategies …
Self-supervised Learning
Definition
Self-supervised learning is a technique where the algorithm creates supervisory signals …
Supervised Fine-tuning
Definition
Supervised Fine-tuning (SFT) involves taking a large pre-trained model, such as a …
Supervised Learning
Definition
In supervised learning, the algorithm is trained on a labeled dataset, meaning each input …
Learning Rate
Definition
The learning rate determines how much the model’s weights are updated relative to …
Loss Function
Definition
Also known as the cost or error function, the loss function provides a scalar value …
Gradient Descent
Definition
Gradient descent is a first-order iterative optimization algorithm for finding a local …
Few-shot Learning
Definition
Few-shot learning aims to enable models to generalize from just a handful of examples, …
fine-tuned
Definition
Fine-tuning involves taking a model that has already been trained on a large, general …
Supervised
Definition
Supervised learning involves feeding an algorithm with data that includes both inputs and …
Pre-training
Definition
Pre-training is a foundational technique in deep learning where a model learns broad …
Loss
Definition
Loss functions, also known as cost functions, measure how well a machine learning …
Instruction Tuning
Definition
This process bridges the gap between general pre-training and specific task performance. …
Fine
Definition
Fine-tuning involves taking a general-purpose model trained on large datasets and further …
Fine-tuning
Definition
Fine-tuning involves taking a model already trained on a large, general dataset and …
Backpropagation
Definition
Backpropagation, short for backward propagation of errors, is a method used in artificial …
Alignment
Definition
Alignment focuses on making sure AI systems do what humans actually want, rather than …