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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ĵĺ

-0.84

 Torah

-0.67

SAS

-0.66

ebus

-0.65

ãĥĨãĤ£

-0.65

è¦ļéĨĴ

-0.64

psc

-0.63

uckland

-0.63

ACTED

-0.62

insula

-0.62

POSITIVE LOGITS

going

0.65

orph

0.65

asca

0.62

iter

0.60

anol

0.60

ins

0.60

arge

0.59

behind

0.59

 peel

0.58

uming

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.