INDEX

Explanations

text related to banners and other forms of advertisement or promotion

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "banner" and its variants

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ulu

-0.76

OTOS

-0.74

urrency

-0.72

cephal

-0.70

nesota

-0.67

olation

-0.64

juries

-0.64

olated

-0.64

iott

-0.63

ivot

-0.62

POSITIVE LOGITS

 banners

1.07

 banner

1.05

pole

1.03

boards

0.82

 proclaiming

0.82

stones

0.81

holder

0.79

ram

0.78

board

0.76

holders

0.76

Activations Density 0.015%

text related to banners and other forms of advertisement or promotion

occurrences of the word "banner" and its variants

No Comments

No Known Activations