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

rite

-0.86

psons

-0.71

nesota

-0.70

baugh

-0.67

hews

-0.66

izabeth

-0.63

char

-0.63

ESP

-0.63

raper

-0.63

atche

-0.63

POSITIVE LOGITS

ËĪ

0.71

PLIED

0.66

andem

0.65

LLOW

0.61

TBA

0.60

Lau

0.58

ãĥĨ

0.58

 closely

0.58

 others

0.58

 neigh

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.