INDEX

Explanations

phrases related to extending or prolonging something

oai_token-act-pair · gpt-3.5-turbo

references to extending agreements or support in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

anooga

-0.72

idd

-0.69

Eng

-0.67

urch

-0.67

atics

-0.67

isky

-0.66

ifferent

-0.64

apest

-0.64

ohn

-0.64

Sty

-0.64

POSITIVE LOGITS

 indefinitely

0.95

 moratorium

0.90

stay

0.89

 warranties

0.84

license

0.79

 bounds

0.79

 privileges

0.77

 extended

0.76

 lifespan

0.75

 warranty

0.74

Activations Density 0.136%

phrases related to extending or prolonging something

references to extending agreements or support in different contexts

No Comments

No Known Activations