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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

«ĺ

-0.66

auld

-0.65

 symb

-0.65

utable

-0.64

»

-0.64

inator

-0.63

´

-0.62

uv

-0.61

ERC

-0.60

ellar

-0.60

POSITIVE LOGITS

essen

0.84

Bei

0.67

 Ride

0.65

ktop

0.63

reens

0.62

 Whites

0.61

 Soldiers

0.61

 Charlottesville

0.61

 Oaks

0.59

 Fors

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.