INDEX

Explanations

references to individuals related to political events or controversies

oai_token-act-pair · gpt-3.5-turbo

names of specific individuals and related entities

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

staff

-0.77

ctl

-0.76

ardless

-0.74

rum

-0.68

Country

-0.67

Tec

-0.67

mint

-0.66

eding

-0.66

glers

-0.66

iencies

-0.64

POSITIVE LOGITS

ovsky

1.02

objects

0.73

 autop

0.71

 Lerner

0.67

yon

0.66

 impe

0.65

oidal

0.63

asury

0.63

aunder

0.61

izabeth

0.61

Activations Density 0.037%

references to individuals related to political events or controversies

names of specific individuals and related entities

No Comments

No Known Activations