INDEX

Explanations

references to legal terms or youth-related crime

oai_token-act-pair · gpt-3.5-turbo

terms related to juvenile individuals and their legal issues

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ding

-0.79

anyahu

-0.75

âķĲ

-0.71

++++++++++++++++

-0.71

cross

-0.69

×Ļ×

-0.69

witz

-0.66

SOURCE

-0.65

ãĤ¡

-0.65

mberg

-0.64

POSITIVE LOGITS

 delinqu

1.43

 delinquent

1.19

 Juven

0.96

 Gohan

0.85

ishly

0.80

 detention

0.79

ety

0.79

 offenders

0.77

 juvenile

0.72

 offender

0.72

Activations Density 0.012%

references to legal terms or youth-related crime

terms related to juvenile individuals and their legal issues

No Comments

No Known Activations