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

merce

-0.77

ulators

-0.76

ath

-0.73

netflix

-0.71

orescence

-0.71

wark

-0.70

ulator

-0.68

onet

-0.68

atron

-0.68

catentry

-0.68

POSITIVE LOGITS

 deducted

0.68

skip

0.67

DUP

0.63

hetical

0.63

bold

0.62

NP

0.62

 srfAttach

0.62

HP

0.62

GES

0.61

 Prevention

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.