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

Buc

-0.73

©¶æ

-0.67

 Logged

-0.65

ĪĴ

-0.64

oÄŁ

-0.62

idation

-0.62

 simul

-0.61

ample

-0.60

acists

-0.60

 decriminal

-0.59

POSITIVE LOGITS

terday

0.75

gallery

0.72

ospace

0.66

 deserve

0.66

IFE

0.64

deen

0.63

iors

0.61

ãĥ³

0.61

killer

0.60

illion

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.