INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

PDATE

-0.85

ciating

-0.77

FORM

-0.76

folios

-0.73

BALL

-0.73

ngth

-0.71

CNN

-0.69

PF

-0.69

DonaldTrump

-0.67

raits

-0.66

POSITIVE LOGITS

 adop

0.67

Ã¶n

0.66

 tails

0.62

 borough

0.62

ÄŁ

0.61

Hos

0.60

 influences

0.59

teness

0.59

Soy

0.58

Mos

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.