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

SI

-0.77

hare

-0.72

cens

-0.72

SF

-0.71

tu

-0.69

HI

-0.68

anth

-0.68

baugh

-0.67

click

-0.64

auld

-0.64

POSITIVE LOGITS

lication

0.75

ftime

0.70

 trailed

0.69

 coron

0.68

phies

0.68

 inherit

0.68

 encount

0.67

chieve

0.66

 coma

0.66

 preceded

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.