INDEX

Explanations

the word component

oai_token-act-pair · gpt-3.5-turbo

references to components or elements in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pants

-0.79

 jails

-0.76

markets

-0.75

facts

-0.75

 lashes

-0.74

olds

-0.72

words

-0.70

 toilets

-0.68

ographs

-0.68

fights

-0.67

POSITIVE LOGITS

 component

3.45

 Component

2.12

component

2.04

 element

1.81

 components

1.81

Component

1.53

 aspect

1.49

 constituent

1.48

 Components

1.42

 ingredient

1.34

Activations Density 0.010%

the word component

references to components or elements in various contexts

No Comments

No Known Activations