INDEX

Explanations

terms related to innocence and victims

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 19-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L19R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.19.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

lfw

-0.17

phan

-0.17

ilet

-0.16

ongan

-0.15

illisecond

-0.15

ionic

-0.15

yte

-0.14

alog

-0.14

imli

-0.14

POSITIVE LOGITS

 innocent

0.27

 bystand

0.25

 innocence

0.24

 innoc

0.24

 Innoc

0.23

 harmless

0.19

 civilians

0.18

/simple

0.17

 victims

0.16

-looking

0.16

Activations Density 0.011%

terms related to innocence and victims

No Comments

No Known Activations