INDEX

Explanations

phrases indicating a component or aspect of a larger entity or event

oai_token-act-pair · gpt-3.5-turbo

references to being involved in activities or initiatives

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

asca

-0.74

 ceilings

-0.73

asu

-0.71

mare

-0.69

 destro

-0.67

 grandchildren

-0.67

akespe

-0.66

ampires

-0.64

avorite

-0.64

liam

-0.64

POSITIVE LOGITS

meal

0.89

displayText

0.77

ials

0.77

ially

0.74

strip

0.70

meier

0.70

aking

0.70

 adjunct

0.68

ridge

0.67

ners

0.67

Activations Density 0.019%

phrases indicating a component or aspect of a larger entity or event

references to being involved in activities or initiatives

No Comments

No Known Activations