INDEX

Explanations

words related to destruction or obliteration

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

Ø§

-0.28

at

-0.23

ÙĩÙħ

-0.23

et

-0.20

Ùħ

-0.20

ine

-0.20

and

-0.19

Ñı

-0.18

lp

-0.18

POSITIVE LOGITS

hape

0.20

zman

0.18

kers

0.17

isle

0.17

alu

0.17

antis

0.17

abeled

0.17

alat

0.17

boro

0.17

hoot

0.17

Activations Density 0.056%

words related to destruction or obliteration

No Comments

No Known Activations