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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Mehran

-0.68

IDENT

-0.67

IVERS

-0.67

CAST

-0.66

NER

-0.65

Reviewer

-0.63

PAR

-0.63

CHAR

-0.63

 befriend

-0.62

 Plot

-0.62

POSITIVE LOGITS

rums

0.70

theless

0.70

lain

0.70

oliath

0.69

arta

0.69

iku

0.68

bm

0.68

thur

0.67

ly

0.67

esty

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.