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

 signings

-0.83

arning

-0.75

arie

-0.72

oki

-0.69

andestine

-0.68

Ĥª

-0.65

lling

-0.65

 Bastard

-0.64

ĻĤ

-0.64

 Nanto

-0.64

POSITIVE LOGITS

IPM

0.62

 streamlined

0.62

Requires

0.61

Graphics

0.61

NV

0.60

xml

0.59

Ref

0.58

Nav

0.57

OTHER

0.57

BG

0.56

Activations Density 0.000%

No Known Activations

This feature has no known activations.