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

ointed

-0.85

soDeliveryDate

-0.81

ordial

-0.77

ratulations

-0.76

onite

-0.74

Ì¶

-0.71

LESS

-0.70

seless

-0.70

 unlocked

-0.68

zed

-0.68

POSITIVE LOGITS

SPONSORED

0.82

Inf

0.69

bytes

0.67

hound

0.67

 Broadcast

0.65

 Derby

0.65

0.64

âĨĳ

0.64

 Burma

0.63

Chel

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.