INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

adjust

-0.79

emort

-0.78

otte

-0.74

aukee

-0.74

natureconservancy

-0.74

affer

-0.73

iors

-0.72

ooth

-0.72

acus

-0.71

rounded

-0.71

POSITIVE LOGITS

 number

1.70

1.05

 Number

1.05

 numbers

0.84

Number

0.81

number

0.81

NUM

0.71

 Numbers

0.70

 numbering

0.69

 hashtag

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.