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

ÙĲ

-0.81

ravings

-0.69

 Pist

-0.67

 silhou

-0.63

 glances

-0.63

bolt

-0.62

wolves

-0.60

wagen

-0.59

Clock

-0.58

angs

-0.58

POSITIVE LOGITS

aeda

0.66

ahime

0.66

cgi

0.65

NIC

0.65

igon

0.64

emort

0.64

achusetts

0.62

manent

0.61

akin

0.60

abad

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.