INDEX

Explanations

phrases related to shooting or capturing images

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

loi

-0.18

cheid

-0.18

riad

-0.17

hire

-0.17

AndPassword

-0.16

hir

-0.16

hud

-0.16

amide

-0.15

ØµØ§Øª

-0.15

chor

-0.15

POSITIVE LOGITS

guns

0.23

gun

0.17

ocoa

0.17

baz

0.16

ermo

0.16

 Rapids

0.16

shots

0.15

Straight

0.15

 Straight

0.15

 shot

0.15

Activations Density 0.028%

phrases related to shooting or capturing images

No Comments

No Known Activations