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

PLA

-0.86

ortment

-0.76

aram

-0.73

ulsion

-0.73

Accessory

-0.73

SU

-0.70

 showc

-0.70

sic

-0.69

Kit

-0.68

are

-0.67

POSITIVE LOGITS

 exagger

0.77

ĻĤ

0.74

 Traps

0.68

yer

0.66

illes

0.65

 Judd

0.64

IGN

0.62

Pwr

0.62

RJ

0.60

 Reynolds

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.