INDEX

Explanations

phrases related to knowledge and understanding

oai_token-act-pair · gpt-3.5-turbo

instances of awareness or understanding

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

cohol

-0.73

atre

-0.73

ermanent

-0.71

clusive

-0.70

bury

-0.70

acco

-0.68

aredevil

-0.68

vity

-0.68

elight

-0.68

ItemTracker

-0.68

POSITIVE LOGITS

ledged

1.14

 instinctively

1.07

how

1.04

ledge

0.99

 firsthand

0.97

 nothing

0.96

 exactly

0.85

 darn

0.84

 what

0.83

nothing

0.83

Activations Density 0.066%

phrases related to knowledge and understanding

instances of awareness or understanding

No Comments

No Known Activations