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

nces

-0.77

cknow

-0.71

 integers

-0.69

 Ranking

-0.65

 âĨĳ

-0.65

packing

-0.64

Py

-0.63

 Clicker

-0.62

 Kazakhstan

-0.62

ordinary

-0.60

POSITIVE LOGITS

bush

0.92

lich

0.77

URA

0.74

assian

0.71

eller

0.69

itta

0.69

ULE

0.65

bed

0.65

VEN

0.65

ideshow

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.