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

thia

-0.88

vertisements

-0.73

disabled

-0.69

gments

-0.68

ascript

-0.66

nant

-0.66

-+-+-+-+

-0.65

atre

-0.65

jad

-0.65

athed

-0.64

POSITIVE LOGITS

 yielding

0.71

 Ring

0.64

 mete

0.63

 Noon

0.63

 afar

0.62

 Dante

0.62

 denomin

0.61

 administering

0.59

venge

0.59

Lot

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.