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

utan

-0.80

osponsors

-0.76

 Daylight

-0.75

natureconservancy

-0.72

 juven

-0.69

 Illum

-0.65

âĪĴ

-0.65

earchers

-0.64

igation

-0.64

ulatory

-0.64

POSITIVE LOGITS

ussen

0.72

gra

0.67

onomy

0.66

rams

0.65

MY

0.62

unker

0.62

ghai

0.62

yi

0.60

ours

0.60

ieu

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.