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

XM

-0.82

 Eater

-0.74

rift

-0.70

SPONSORED

-0.68

stem

-0.66

HER

-0.65

gue

-0.64

stuff

-0.64

iverse

-0.64

Frag

-0.63

POSITIVE LOGITS

ancies

0.84

ental

0.63

 lockout

0.63

ourses

0.63

ancy

0.62

 duties

0.62

 watering

0.61

eeds

0.60

uations

0.60

 Mehran

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.