Input Representation in a Transformer Layer
The input to a Transformer layer at a given depth is represented as a sequence of hidden states, denoted by . In this notation, is the sequence containing all hidden states for the layer.
0
1
Tags
Ch.3 Prompting - Foundations of Large Language Models
Foundations of Large Language Models
Foundations of Large Language Models Course
Computing Sciences
Related
Input Representation in a Transformer Layer
Comparison of Prompt Tuning and Prefix Fine-Tuning
Input Composition in a Prefix-Tuned Transformer Layer
A research team is adapting a pre-trained language model for a specialized legal document summarization task. To conserve computational resources, they decide against retraining the entire model. Instead, for each layer of the model's architecture, they introduce a small set of new, trainable vectors. These vectors are prepended to the sequence of hidden states that serve as input for that layer. During training, only these newly introduced vectors are updated, while the original model parameters are kept frozen. Which statement accurately analyzes the team's approach?
Evaluating a Parameter-Efficient Tuning Method
Efficiency of Prefix Fine-Tuning
Architectural Preservation by Separating Soft Prompts from LLMs
A development team is adapting a large language model for a new task using a method where they freeze all original model weights. For each layer in the model, they prepend a small, unique sequence of trainable vectors to that layer's input. Based on this description, which statement best evaluates the primary trade-off of this technique?
Your team is building a multi-tenant LLM service w...
You’re reviewing an internal design doc for adapti...
You’re implementing a PEFT approach for a customer...
You’re reviewing a teammate’s claim about a new PE...
Diagnosing a PEFT Implementation Bug: Prompt Tuning vs Prefix Fine-Tuning
Choosing and Explaining a PEFT Strategy Under Deployment Constraints
Selecting Prompt Tuning vs Prefix Fine-Tuning by Reasoning from Where Soft Prompts Enter the Transformer
Post-Deployment PEFT Choice and Prefix Input Composition for a Multi-Tenant LLM Service
Choosing Between Prompt Tuning and Prefix Fine-Tuning for a Latency-Critical, Multi-Task LLM Service
Root-Causing a Prefix-Tuning Rollout Regression in a Multi-Task LLM Platform
Input Representation in a Transformer Layer
A computational model is processing a sequence of three items (at positions 0, 1, and 2). The model's internal state at its 5th processing stage is represented by the collection of vectors . What does the single vector represent in this context?
Representing a Layer's State for a Sequence
Applying Layer State Representation
Learn After
Transformer Layer Output Formula
General Formula for a Transformer Layer
Input Composition in a Prefix-Tuned Transformer Layer
A language model is processing an input sentence that has been broken down into 5 distinct tokens. The input to the first processing layer is represented as a matrix containing 5 separate vectors, one for each token. Why is it fundamentally important for the model to maintain this structure—a sequence of individual vectors—as the input to each subsequent layer, rather than, for example, averaging or concatenating them into a single vector?
Structure of a Transformer Layer's Input
When a Transformer model processes a sentence with 12 tokens, the input to the fifth layer is a single, high-dimensional vector that represents the aggregated meaning of the entire sentence as computed by the first four layers.