INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ï¸ı

-0.94

htt

-0.79

LESS

-0.79

INST

-0.77

ÙĲ

-0.75

WARE

-0.75

CAST

-0.75

å¾

-0.74

CHO

-0.74

SPEC

-0.73

POSITIVE LOGITS

isl

0.70

amba

0.68

ammy

0.68

 eyebrows

0.67

 Thieves

0.66

inki

0.66

ked

0.66

 penis

0.63

 redund

0.62

 Jaguars

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.