INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

iP

-0.68

Anth

-0.67

APD

-0.66

fell

-0.65

BP

-0.65

vd

-0.62

mpeg

-0.62

Shift

-0.60

mp

-0.60

POSITIVE LOGITS

egal

0.78

swer

0.69

mediate

0.67

edience

0.66

vantage

0.66

advant

0.63

atable

0.62

sembly

0.62

 Installation

0.60

usting

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.