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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hower

-0.75

fitting

-0.73

CLASSIFIED

-0.71

ossibility

-0.69

minist

-0.66

anism

-0.64

GBT

-0.62

administ

-0.61

 ceremonial

-0.61

owitz

-0.61

POSITIVE LOGITS

ãĤ´ãĥ³

0.83

apolis

0.73

CLOSE

0.69

ukong

0.67

yne

0.67

 Tackle

0.64

 Soda

0.63

Pep

0.63

asta

0.62

 brace

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.