INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 scattered

-0.67

 uniforms

-0.60

 inconsistent

-0.60

 exchanges

-0.59

 offensive

-0.59

 practise

-0.58

 perimeter

-0.58

 practice

-0.57

 caval

-0.56

BC

-0.56

POSITIVE LOGITS

nect

0.82

mma

0.79

oÄŁ

0.79

hiro

0.79

yss

0.75

lev

0.72

igator

0.71

redits

0.70

pez

0.69

Drag

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.