INDEX

Explanations

mentions of prisons or related terms like prison sentences

oai_token-act-pair · gpt-3.5-turbo

references to prisons and the prison system

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

yip

-0.95

lass

-0.79

ï¸ı

-0.79

thora

-0.77

udden

-0.74

omatic

-0.71

rians

-0.70

rian

-0.69

::::::::

-0.68

witz

-0.67

POSITIVE LOGITS

 inmates

0.93

 prisons

0.86

 inmate

0.84

 prison

0.82

prison

0.80

 barr

0.79

 sentences

0.78

gul

0.78

 confinement

0.76

 camps

0.75

Activations Density 0.026%

mentions of prisons or related terms like prison sentences

references to prisons and the prison system

No Comments

No Known Activations