INDEX

Explanations

phrases referring to locations or positions, like "in" or "as"

oai_token-act-pair · gpt-3.5-turbo

phrases indicating inclusion or participation 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 @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tein

-0.69

axy

-0.68

anu

-0.65

emo

-0.64

adr

-0.64

undy

-0.62

atism

-0.62

planet

-0.62

arth

-0.61

inous

-0.60

POSITIVE LOGITS

Magikarp

0.81

ISON

0.69

 Liberties

0.64

lance

0.59

 regards

0.59

sylv

0.58

pers

0.57

interstitial

0.57

å£«

0.55

ãĥĥãĥī

0.55

Activations Density 0.626%

phrases referring to locations or positions, like "in" or "as"

phrases indicating inclusion or participation in various contexts

No Comments

No Known Activations