INDEX

Explanations

phrases indicating a high degree of quality or significance

oai_token-act-pair · gpt-3.5-turbo

phrases indicating strong opinions or definitive statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

artments

-0.68

eral

-0.68

ifferent

-0.65

istries

-0.62

osponsors

-0.61

urches

-0.60

yrics

-0.59

yss

-0.59

ibilities

-0.59

iries

-0.59

POSITIVE LOGITS

 underrated

1.03

THE

0.95

the

0.87

one

0.86

 arguably

0.84

 unmatched

0.83

 coolest

0.81

 unparalleled

0.80

the

0.79

 best

0.78

Activations Density 0.354%

phrases indicating a high degree of quality or significance

phrases indicating strong opinions or definitive statements

No Comments

No Known Activations