INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

SPONSORED

-0.88

OPLE

-0.82

-0.64

OTA

-0.64

————

-0.64

externalActionCode

-0.63

=""

-0.63

CHR

-0.63

bec

-0.62

 isEnabled

-0.61

POSITIVE LOGITS

ingly

1.66

ings

0.96

ing

0.89

ozy

0.80

sburgh

0.78

ishly

0.77

ingham

0.73

esome

0.69

teness

0.69

asca

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.