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

oggle

-0.81

thood

-0.79

ilde

-0.79

ards

-0.76

alloween

-0.67

ertodd

-0.66

ymes

-0.64

 likened

-0.64

ayne

-0.64

isec

-0.63

POSITIVE LOGITS

Unity

0.71

actionDate

0.70

lio

0.66

 yourselves

0.64

upt

0.63

jri

0.61

activity

0.60

 Rising

0.59

 anticipation

0.58

 srfAttach

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.