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

aun

-0.70

 recycle

-0.69

 maiden

-0.68

pless

-0.68

atown

-0.62

Tai

-0.62

 Splash

-0.61

ibaba

-0.60

acea

-0.59

quartered

-0.58

POSITIVE LOGITS

IU

0.74

onom

0.74

Newsletter

0.67

Os

0.67

xon

0.66

IME

0.66

pec

0.64

nesota

0.63

rones

0.62

Sep

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.