INDEX

Explanations

mentions or references to presence or absence

oai_token-act-pair · gpt-3.5-turbo

mentions of "presence"

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

imb

-0.85

initions

-0.70

cipled

-0.69

bard

-0.69

ifiable

-0.68

DIT

-0.67

aired

-0.66

endiary

-0.65

YR

-0.65

strap

-0.64

POSITIVE LOGITS

 presence

1.07

 Presence

0.87

enance

0.78

uated

0.75

idon

0.72

terday

0.68

luster

0.67

 abroad

0.66

alan

0.65

uality

0.65

Activations Density 0.019%

mentions or references to presence or absence

mentions of "presence"

No Comments

No Known Activations