INDEX

Explanations

mentions or prompts related to social media platforms, particularly Google and Google Play Store

oai_token-act-pair · gpt-3.5-turbo

mentions of the term "Google" and its associated services

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

iov

-0.71

ento

-0.70

osta

-0.69

akis

-0.68

mble

-0.67

minster

-0.65

ships

-0.65

icka

-0.63

sole

-0.62

qv

-0.61

POSITIVE LOGITS

 Scholar

1.05

+.

0.99

+,

0.94

0.92

 Maps

0.88

maps

0.88

usercontent

0.87

Play

0.87

+)

0.86

 Play

0.82

Activations Density 0.023%

mentions or prompts related to social media platforms, particularly Google and Google Play Store

mentions of the term "Google" and its associated services

No Comments

No Known Activations