INDEX

Explanations

No Explanations Found

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

isconsin

-0.68

ãĥĬ

-0.66

icion

-0.64

 frivol

-0.63

interstitial

-0.63

 Dian

-0.61

cler

-0.60

Mub

-0.60

ggies

-0.60

entious

-0.58

POSITIVE LOGITS

urated

0.69

kefeller

0.64

oa

0.61

taker

0.61

acteria

0.61

ole

0.61

itute

0.60

 shortcut

0.60

 Organisation

0.59

ttp

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.