INDEX

Explanations

numeric quantities related to count or volume

oai_token-act-pair · gpt-3.5-turbo

mentions of items, concepts, or entities that are included in a discussion or report

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Sabha

-0.77

ickr

-0.65

INESS

-0.65

?]

-0.64

 Baird

-0.64

VM

-0.63

ASA

-0.62

imeo

-0.61

assetsadobe

-0.61

Sport

-0.59

POSITIVE LOGITS

ographic

0.74

 prominently

0.72

 spoilers

0.70

osher

0.69

rawdownloadcloneembedreportprint

0.69

 elements

0.68

 tantal

0.66

ographical

0.66

otyp

0.63

 signature

0.62

Activations Density 0.516%

numeric quantities related to count or volume

mentions of items, concepts, or entities that are included in a discussion or report

No Comments

No Known Activations