INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 scrut

-0.85

verty

-0.79

IFA

-0.76

earch

-0.73

ionage

-0.72

 Ethiop

-0.72

 fundament

-0.72

FTWARE

-0.69

 unlaw

-0.67

ILL

-0.66

POSITIVE LOGITS

ga

0.77

nation

0.72

god

0.70

flo

0.67

reb

0.65

 knock

0.64

atl

0.64

nam

0.64

gan

0.63

atra

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.