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

reviewed

-0.70

hower

-0.67

iversary

-0.67

kefeller

-0.63

stros

-0.63

collection

-0.62

oran

-0.62

chnology

-0.61

odox

-0.61

 pardon

-0.61

POSITIVE LOGITS

NetMessage

0.97

PsyNetMessage

0.83

AMS

0.78

teen

0.78

chy

0.74

Generic

0.72

OUND

0.71

pmwiki

0.70

MSN

0.70

)].

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.