INDEX

Explanations

references to something being "hot" or related to warmth or popularity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ignon

-0.17

ingen

-0.16

ately

-0.16

ean

-0.16

alloc

-0.15

hy

-0.15

hower

-0.15

hari

-0.15

histor

-0.15

nech

-0.15

POSITIVE LOGITS

spots

0.38

shot

0.34

dog

0.33

chk

0.31

dogs

0.31

shots

0.31

bed

0.31

spot

0.30

ter

0.29

rod

0.27

Activations Density 0.019%

references to something being "hot" or related to warmth or popularity

No Comments

No Known Activations