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

 Aires

-0.73

stars

-0.67

 Emin

-0.67

 Petra

-0.66

-0.63

El

-0.62

El

-0.62

é¾įåĸļå£«

-0.61

 Lies

-0.61

 Olympia

-0.61

POSITIVE LOGITS

?]

0.74

ixon

0.70

 carrier

0.69

tailed

0.68

 exception

0.67

arov

0.64

arin

0.63

omatic

0.63

etooth

0.63

anza

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.