INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ulton

-0.62

nos

-0.62

 Frankfurt

-0.61

 Kendall

-0.60

 traps

-0.60

 Known

-0.60

Chr

-0.59

wick

-0.59

aline

-0.59

esses

-0.59

POSITIVE LOGITS

¬¼

0.80

monary

0.80

ÄŁ

0.77

jam

0.74

ãĥ¼ãĥ³

0.74

lder

0.68

arij

0.66

nesday

0.66

åī

0.65

 landslide

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.