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

abouts

-0.71

itia

-0.69

ãĤ©

-0.66

ãĤ³

-0.64

 Serving

-0.64

ãĥīãĥ©ãĤ´ãĥ³

-0.64

bable

-0.62

things

-0.61

backs

-0.61

ãĥĥãĥī

-0.61

POSITIVE LOGITS

lies

0.68

Corp

0.65

 Hung

0.64

ui

0.63

angler

0.63

star

0.63

bian

0.63

Shi

0.62

immer

0.62

 transmission

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.