INDEX

Explanations

elements related to software or user interface design, particularly focusing on toolbar elements

oai_token-act-pair · gpt-3.5-turbo

elements related to user interface components and their functionalities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

fund

-0.83

istani

-0.79

cius

-0.78

anol

-0.78

breeding

-0.77

ffer

-0.73

angan

-0.73

oil

-0.72

ivers

-0.71

VEN

-0.70

POSITIVE LOGITS

 toolbar

1.15

 widget

1.04

 layout

0.97

 widgets

0.96

 interface

0.95

 scrolling

0.95

 Layout

0.94

 navigation

0.92

UI

0.92

 popup

0.91

Activations Density 0.171%

elements related to software or user interface design, particularly focusing on toolbar elements

elements related to user interface components and their functionalities

No Comments

No Known Activations