INDEX

Explanations

phrases where accountability is assigned, particularly related to law enforcement

oai_token-act-pair · gpt-3.5-turbo

possessive forms indicating ownership or association with entities

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

lehem

-0.82

jun

-0.78

olkien

-0.75

lin

-0.75

inet

-0.75

unin

-0.73

hari

-0.73

ubuntu

-0.71

istrate

-0.71

obin

-0.71

POSITIVE LOGITS

own

1.26

 newest

1.04

 foray

1.03

 insistence

1.02

 penchant

1.01

 inability

0.99

 plight

0.99

 predicament

0.96

 unwillingness

0.93

 newfound

0.93

Activations Density 0.169%

phrases where accountability is assigned, particularly related to law enforcement

possessive forms indicating ownership or association with entities

No Comments

No Known Activations