INDEX

Explanations

the word "unfamiliar" or related variations

oai_token-act-pair · gpt-3.5-turbo

terms related to unfamiliarity and inexperience

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

ramid

-0.92

owder

-0.77

raq

-0.74

die

-0.73

trak

-0.72

roxy

-0.71

essen

-0.67

di

-0.67

aminer

-0.67

illance

-0.67

POSITIVE LOGITS

 unfamiliar

1.38

ity

0.88

 familiar

0.77

lihood

0.75

 arous

0.75

uously

0.74

 acquaintance

0.72

sworth

0.68

etheless

0.67

 strangers

0.67

Activations Density 0.007%

the word "unfamiliar" or related variations

terms related to unfamiliarity and inexperience

No Comments

No Known Activations