INDEX

Explanations

references to Google Maps

oai_token-act-pair · gpt-3.5-turbo

mentions of the Google Maps application

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

inion

-0.93

nder

-0.85

pton

-0.79

tail

-0.77

isher

-0.74

ned

-0.73

nen

-0.73

complex

-0.72

Thumbnail

-0.71

inity

-0.71

POSITIVE LOGITS

 Maps

1.17

 maps

1.10

Maps

0.93

maps

0.90

map

0.89

 mapped

0.89

 mapping

0.83

ynthesis

0.76

ynt

0.75

Map

0.74

Activations Density 0.007%

references to Google Maps

mentions of the Google Maps application

No Comments

No Known Activations