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

 Homs

-0.77

cery

-0.73

@#&

-0.69

ptin

-0.68

amental

-0.66

`,

-0.66

cised

-0.66

mounted

-0.65

 Assassin

-0.65

tics

-0.64

POSITIVE LOGITS

Typ

0.69

arsity

0.67

aday

0.66

antic

0.66

 penn

0.65

 newsp

0.65

 tiers

0.64

 arrang

0.62

Cla

0.62

Spons

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.