INDEX

Explanations

technical features or attributes of various products or services

oai_token-act-pair · gpt-3.5-turbo

keywords related to objects, items, and entities commonly referenced in discussions or reviews

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

iencies

-0.75

CLR

-0.64

 parties

-0.64

 desks

-0.64

 Doors

-0.60

 Sections

-0.59

 automakers

-0.59

 sides

-0.59

Bir

-0.58

 Roof

-0.58

POSITIVE LOGITS

cture

0.79

oster

0.69

folio

0.69

eless

0.65

ittee

0.65

 belongs

0.65

stros

0.65

iece

0.65

isphere

0.63

ourney

0.62

Activations Density 0.301%

technical features or attributes of various products or services

keywords related to objects, items, and entities commonly referenced in discussions or reviews

No Comments

No Known Activations