INDEX

Explanations

words related to specific geographical locations, particularly cities

oai_token-act-pair · gpt-3.5-turbo

mentions of cities

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

igham

-0.97

imilar

-0.81

opsis

-0.80

etsk

-0.80

chnology

-0.80

sembly

-0.78

arlane

-0.78

wcsstore

-0.76

zees

-0.76

terday

-0.76

POSITIVE LOGITS

scape

1.28

wide

1.00

 skyline

0.95

 dwell

0.94

 councillor

0.87

 streets

0.85

 hall

0.85

 council

0.82

 halls

0.81

 councillors

0.80

Activations Density 0.049%

words related to specific geographical locations, particularly cities

mentions of cities

No Comments

No Known Activations