INDEX

Explanations

JavaScript-related strings that prompt enabling or using JavaScript in web contexts

oai_token-act-pair · gpt-3.5-turbo

references to JavaScript programming language

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rower

-0.77

naires

-0.77

ially

-0.75

llah

-0.74

 Cathy

-0.65

Ĥª

-0.65

ŃĶ

-0.62

 Manz

-0.62

Kub

-0.61

ership

-0.60

POSITIVE LOGITS

ESSION

0.82

 Enabled

0.78

 enabled

0.78

 Babel

0.77

 interpreter

0.73

JS

0.71

 Detected

0.69

Unit

0.69

 JavaScript

0.69

avascript

0.66

Activations Density 0.020%

JavaScript-related strings that prompt enabling or using JavaScript in web contexts

references to JavaScript programming language

No Comments

No Known Activations