INDEX

Explanations

mentions of legal actions or criminal activities in news articles

oai_token-act-pair · gpt-3.5-turbo

words related to criminal activities and legal charges

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

pees

-0.77

zek

-0.67

ynes

-0.66

 horizont

-0.64

untled

-0.64

 challeng

-0.63

 stagn

-0.63

 intrins

-0.63

bley

-0.62

orsi

-0.62

POSITIVE LOGITS

 Roland

0.71

 Diplom

0.68

was

0.65

sets

0.65

Ai

0.65

 Tata

0.64

 Conan

0.62

ourke

0.62

ages

0.60

 Kathryn

0.60

Activations Density 1.141%

mentions of legal actions or criminal activities in news articles

words related to criminal activities and legal charges

No Comments

No Known Activations