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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

azon

-0.80

icates

-0.79

 Tanzania

-0.76

luaj

-0.75

ĸļ

-0.75

icators

-0.75

berra

-0.73

 hooks

-0.71

ihad

-0.71

mes

-0.71

POSITIVE LOGITS

Deal

0.79

Ger

0.69

ENA

0.68

 Hail

0.67

External

0.66

Drive

0.65

VP

0.65

Destroy

0.65

Vul

0.64

deg

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.