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

©¶æ

-0.77

owsky

-0.73

axter

-0.72

obin

-0.69

omatic

-0.67

 Bauer

-0.67

omore

-0.66

ĸļ

-0.65

tis

-0.65

 Michaels

-0.64

POSITIVE LOGITS

 unfocusedRange

0.80

 Lank

0.69

Ø§

0.66

DEC

0.66

fighting

0.65

govtrack

0.62

arnaev

0.61

pees

0.60

ructose

0.59

ANC

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.