INDEX

Explanations

phrases related to public opinion or voting outcomes

oai_token-act-pair · gpt-3.5-turbo

references to the concept of popularity in various contexts

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

thur

-0.75

ourke

-0.70

 Aviv

-0.70

abetic

-0.68

 Kear

-0.67

agher

-0.67

xual

-0.66

ritch

-0.64

 Territ

-0.63

ASC

-0.63

POSITIVE LOGITS

ized

0.96

izing

0.92

isations

0.91

izations

0.89

ity

0.87

ised

0.86

ization

0.76

izer

0.76

ize

0.74

izers

0.73

Activations Density 0.014%

phrases related to public opinion or voting outcomes

references to the concept of popularity in various contexts

No Comments

No Known Activations