INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

-0.72

ival

-0.71

idas

-0.71

netflix

-0.68

-0.67

ById

-0.67

qqa

-0.67

apter

-0.66

otos

-0.65

Wave

-0.65

POSITIVE LOGITS

 Leadership

0.68

 Immigration

0.67

 Challenges

0.64

 challenges

0.64

 inexper

0.64

 architecture

0.63

 Chall

0.62

 constitution

0.60

 Personality

0.60

 Fork

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.