INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ALT

-0.69

 LIMITED

-0.68

ECO

-0.67

\":

-0.67

 Indra

-0.66

 chefs

-0.62

Prof

-0.62

bart

-0.60

 Laksh

-0.60

 Eugene

-0.59

POSITIVE LOGITS

wrapper

0.70

ombie

0.69

iasis

0.68

igslist

0.67

watching

0.67

keleton

0.66

enegger

0.65

 Lovecraft

0.63

webkit

0.63

veyard

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.