INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

hammad

-0.84

anza

-0.72

thur

-0.70

sonian

-0.69

amar

-0.68

clusions

-0.67

ysis

-0.66

itially

-0.66

hawks

-0.66

 loophole

-0.66

POSITIVE LOGITS

Magazine

0.70

 Pont

0.69

 wedd

0.67

ennes

0.67

wcsstore

0.63

enza

0.61

Ed

0.60

iw

0.59

IB

0.59

AMD

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.