INDEX

Explanations

references to a physical bar or related settings

oai_token-act-pair · gpt-3.5-turbo

references to bars in various contexts

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

lihood

-0.94

UAL

-0.86

IBLE

-0.81

wip

-0.70

sie

-0.65

ãģį

-0.65

ibilities

-0.62

CVE

-0.62

ATK

-0.61

 Dawkins

-0.60

POSITIVE LOGITS

itone

1.29

celona

1.27

bell

1.18

iatric

1.12

bers

1.08

raged

1.07

becue

1.04

onial

1.03

bara

0.99

code

0.99

Activations Density 0.029%

references to a physical bar or related settings

references to bars in various contexts

No Comments

No Known Activations