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

bender

-0.80

opian

-0.72

entity

-0.69

orio

-0.69

ERO

-0.67

gression

-0.67

ombat

-0.66

arel

-0.66

angelo

-0.66

Poe

-0.66

POSITIVE LOGITS

wikipedia

0.75

 alike

0.74

etc

0.66

--------------------------------------------------------

0.65

google

0.64

 administrations

0.64

cies

0.63

 caches

0.63

/-

0.62

 taxpayers

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.