INDEX

Explanations

references to indoor activities or settings

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

naires

-0.16

dio

-0.15

addon

-0.15

idel

-0.14

 tÆ°

-0.14

ioc

-0.14

 outer

-0.14

.scalablytyped

-0.14

note

-0.13

POSITIVE LOGITS

/out

0.33

-out

0.27

-Out

0.24

warf

0.17

ountain

0.17

izzo

0.16

/email

0.15

outs

0.15

oop

0.15

essen

0.15

Activations Density 0.005%

references to indoor activities or settings

No Comments

No Known Activations