INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

nered

-0.78

owan

-0.75

geries

-0.75

eele

-0.74

abase

-0.71

ners

-0.70

heny

-0.70

awa

-0.69

uggets

-0.67

oppable

-0.67

POSITIVE LOGITS

helm

0.82

 sacrific

0.74

 reluct

0.71

 explan

0.70

yss

0.68

awa

0.68

 palp

0.66

 conduc

0.66

wind

0.64

witch

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.