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

tic

-0.86

umed

-0.76

rit

-0.75

ennes

-0.70

ich

-0.69

istani

-0.69

ilogy

-0.68

bind

-0.68

stakes

-0.67

worms

-0.66

POSITIVE LOGITS

 Pasadena

0.82

 Goddard

0.78

 Pearce

0.75

SPONSORED

0.74

 Hayward

0.72

 astronomer

0.71

 Berman

0.70

 Boyle

0.68

,,,,

0.67

 Bucc

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.