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

»Ĵ

-1.02

urden

-0.80

psons

-0.76

amaz

-0.71

ereo

-0.69

ammy

-0.69

anamo

-0.69

olphin

-0.67

FILE

-0.67

irlfriend

-0.66

POSITIVE LOGITS

parable

0.73

cially

0.71

Lur

0.69

 Gork

0.65

 decl

0.63

 Ware

0.61

 WARN

0.61

fields

0.61

 Serge

0.60

 hyper

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.