INDEX

Explanations

keywords related to a specific website or publication, in this case 'Business Insider'

oai_token-act-pair · gpt-3.5-turbo

mentions of the term "Insider"

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

itism

-0.83

apy

-0.82

ite

-0.79

ites

-0.78

IZ

-0.74

iaz

-0.74

oiler

-0.71

itions

-0.71

ifting

-0.71

dogs

-0.71

POSITIVE LOGITS

REDACTED

0.69

 Trading

0.69

OPLE

0.67

Aus

0.65

 Cousins

0.64

undy

0.64

agy

0.64

 Commander

0.62

 Eater

0.61

querque

0.61

Activations Density 0.030%

keywords related to a specific website or publication, in this case 'Business Insider'

mentions of the term "Insider"

No Comments

No Known Activations