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

rongh

-0.86

bucks

-0.76

assian

-0.76

HAM

-0.76

miah

-0.75

yton

-0.74

achy

-0.72

ippi

-0.72

nces

-0.70

boro

-0.69

POSITIVE LOGITS

 sear

0.70

 lodge

0.68

 transgress

0.67

tet

0.64

 metals

0.64

 recipients

0.63

Sat

0.62

 disse

0.61

 virt

0.60

 deposit

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.