INDEX

Explanations

words or phrases indicating doubt or challenge

oai_token-act-pair · gpt-3.5-turbo

instances of the word "question" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

rites

-0.86

oiler

-0.74

osponsors

-0.70

\\\\

-0.68

emetery

-0.67

ategory

-0.67

alions

-0.66

 teasp

-0.66

ëĭ

-0.66

Torrent

-0.66

POSITIVE LOGITS

 whether

1.08

naires

1.06

why

0.98

 motives

0.94

whether

0.84

ibly

0.83

ably

0.80

ingly

0.77

naire

0.77

how

0.77

Activations Density 0.067%

words or phrases indicating doubt or challenge

instances of the word "question" and its variations

No Comments

No Known Activations