INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ACTIONS

-0.73

 Garr

-0.70

Ow

-0.67

 Synd

-0.65

Steven

-0.63

otions

-0.61

Iw

-0.61

 Spons

-0.61

aru

-0.60

 Serving

-0.60

POSITIVE LOGITS

 pestic

0.74

sterdam

0.70

orem

0.66

icultural

0.64

 manure

0.64

Alternative

0.63

atile

0.62

export

0.62

 bunker

0.62

nun

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.