INDEX

Explanations

references to specific names, possibly related to locations or individuals

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names associated with specific locations and individuals

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ographed

-0.91

ahon

-0.85

ograp

-0.73

ucket

-0.71

chnology

-0.69

arser

-0.67

oses

-0.67

Ģ

-0.66

ierrez

-0.65

ople

-0.63

POSITIVE LOGITS

shire

0.82

anke

0.78

tyard

0.77

cru

0.75

ding

0.73

ended

0.73

marked

0.73

idav

0.73

stein

0.72

gence

0.72

Activations Density 0.089%

references to specific names, possibly related to locations or individuals

proper nouns, particularly names associated with specific locations and individuals

No Comments

No Known Activations