INDEX

Explanations

mentions of wet conditions or related terms

oai_token-act-pair · gpt-3.5-turbo

references to wet weather or conditions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

å£«

-0.74

xual

-0.67

########

-0.66

vernment

-0.65

OSH

-0.64

 distingu

-0.64

CVE

-0.63

ZA

-0.63

pta

-0.62

inic

-0.61

POSITIVE LOGITS

ting

1.22

ter

1.15

ted

1.09

ness

1.02

bag

0.92

ters

0.92

nes

0.92

nesday

0.91

lands

0.91

 nood

0.90

Activations Density 0.029%

mentions of wet conditions or related terms

references to wet weather or conditions

No Comments

No Known Activations