INDEX

Explanations

phrases or sentences indicating disagreement

oai_token-act-pair · gpt-3.5-turbo

instances of disagreement or dissenting opinions

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

amina

-0.79

GV

-0.74

oufl

-0.69

 Roads

-0.69

spring

-0.68

 Jackets

-0.65

maximum

-0.64

 adrenaline

-0.62

Ãł

-0.62

uxe

-0.61

POSITIVE LOGITS

rences

0.88

ially

0.85

 vehemently

0.82

 disagree

0.81

uously

0.78

edIn

0.78

lihood

0.76

llah

0.71

atively

0.71

uous

0.71

Activations Density 0.026%

phrases or sentences indicating disagreement

instances of disagreement or dissenting opinions

No Comments

No Known Activations