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

actionDate

-0.79

¥µ

-0.76

é¾įåĸļå£«

-0.75

itters

-0.74

umbn

-0.73

isite

-0.72

bers

-0.70

endi

-0.70

itted

-0.66

ait

-0.65

POSITIVE LOGITS

 Franch

0.95

 Corpor

0.66

 Solo

0.63

Saf

0.63

 Rover

0.59

pmwiki

0.59

 regrett

0.59

 Stre

0.59

 royalty

0.59

 Pipeline

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.