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

Khe

-0.69

²¾

-0.68

ativity

-0.68

Naz

-0.67

 acad

-0.67

AGE

-0.64

 Ober

-0.63

 Hunts

-0.61

 Kinnikuman

-0.61

Rox

-0.61

POSITIVE LOGITS

YE

0.76

hide

0.73

 msec

0.73

Interstitial

0.69

hetti

0.68

ividual

0.67

=\"

0.67

unker

0.66

tein

0.66

 bumps

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.