INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 jams

-0.63

 Tumblr

-0.63

 ident

-0.62

jam

-0.60

 default

-0.59

 exclaim

-0.58

cry

-0.57

 flies

-0.56

 Airbus

-0.56

 clocks

-0.56

POSITIVE LOGITS

ternity

0.88

tarian

0.83

hani

0.80

iversal

0.79

abetes

0.78

rama

0.78

merce

0.76

pport

0.76

ĪĴ

0.76

asper

0.76

Activations Density 0.000%

No Known Activations

This feature has no known activations.