INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

PsyNetMessage

-0.78

Bott

-0.71

EPA

-0.67

UI

-0.66

 enthusi

-0.65

NPR

-0.64

RF

-0.63

NetMessage

-0.62

 Reuters

-0.62

govtrack

-0.60

POSITIVE LOGITS

ocene

0.87

imaru

0.76

ricular

0.74

qus

0.73

thinkable

0.71

hab

0.70

ileged

0.70

opathic

0.69

 Rept

0.66

 Chand

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.