INDEX

Explanations

phrases related to global locations or regions

oai_token-act-pair · gpt-3.5-turbo

references to locations or contexts related to various places and communities

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

xual

-0.79

inen

-0.72

staking

-0.65

 Tigers

-0.64

ysis

-0.64

qua

-0.63

etts

-0.63

BT

-0.62

 tatt

-0.61

nis

-0.60

POSITIVE LOGITS

 corners

0.85

abouts

0.85

eatures

0.80

clock

0.77

unin

0.72

perty

0.70

«

0.64

rend

0.64

world

0.64

=~=~

0.64

Activations Density 0.046%

phrases related to global locations or regions

references to locations or contexts related to various places and communities

No Comments

No Known Activations