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

SQU

-0.79

uzz

-0.74

 Bunny

-0.71

 GOODMAN

-0.71

RP

-0.69

 DEFENSE

-0.65

FW

-0.65

 Remastered

-0.65

 sshd

-0.64

HIP

-0.63

POSITIVE LOGITS

inent

0.73

lasting

0.68

icago

0.66

Fig

0.66

icularly

0.64

 territ

0.64

foreign

0.62

 centr

0.62

cia

0.62

erey

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.