INDEX

Explanations

specific mentions of the word "banner"

oai_token-act-pair · gpt-3.5-turbo

references to banners and their significance 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 @ 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

urrency

-0.77

ulu

-0.75

ube

-0.74

AE

-0.74

cephal

-0.73

lege

-0.71

nesota

-0.67

OTOS

-0.66

Tob

-0.66

erm

-0.64

POSITIVE LOGITS

 banner

0.96

 banners

0.94

holder

0.92

 proclaiming

0.89

boards

0.86

holders

0.80

stones

0.80

board

0.79

strip

0.79

pole

0.77

Activations Density 0.048%

specific mentions of the word "banner"

references to banners and their significance in various contexts

No Comments

No Known Activations