INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

Tax

-0.79

Interstitial

-0.76

Enhanced

-0.75

externalActionCode

-0.73

Miss

-0.70

Premium

-0.69

Fish

-0.68

Contract

-0.68

boats

-0.67

Support

-0.67

POSITIVE LOGITS

ategory

0.78

abin

0.66

ibaba

0.64

iblical

0.64

heny

0.64

chwitz

0.64

userc

0.63

izzard

0.63

ylene

0.63

glomer

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.