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

anguage

-0.75

mble

-0.71

 Belgian

-0.70

eware

-0.68

ozyg

-0.68

stanbul

-0.67

olics

-0.67

 SOFTWARE

-0.67

 Fathers

-0.66

 Mosque

-0.66

POSITIVE LOGITS

IAS

0.72

jab

0.71

otes

0.69

ross

0.69

destruct

0.68

 Morales

0.67

WARD

0.67

Dick

0.66

escape

0.66

Jones

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.