INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

¶ħ

-0.92

 exerc

-0.79

EStreamFrame

-0.69

 adip

-0.68

ŃĶ

-0.67

ĸļ

-0.66

isode

-0.65

sul

-0.64

anguage

-0.64

 unsolved

-0.62

POSITIVE LOGITS

Leaks

0.98

Studio

0.72

Bow

0.71

bugs

0.67

CHAT

0.67

Beck

0.66

standard

0.65

ORN

0.65

letters

0.63

Shock

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.