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

heny

-0.83

rogen

-0.79

Ly

-0.76

XY

-0.74

arks

-0.74

ĸļ

-0.74

hent

-0.74

wcs

-0.73

phen

-0.73

ronics

-0.73

POSITIVE LOGITS

 sacrific

0.74

Pru

0.73

 Krishna

0.72

 fodder

0.70

pad

0.69

rom

0.69

 Nero

0.67

 vine

0.67

 premises

0.66

 vain

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.