INDEX

Explanations

phrases related to national interests and political objectives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

.flink

-0.15

æŀĲ

-0.15

ora

-0.14

oli

-0.14

out

-0.14

rote

-0.14

Ãª

-0.14

erea

-0.14

 Bard

-0.14

çģ£

-0.14

POSITIVE LOGITS

 objectives

0.17

 agendas

0.15

 goals

0.15

ÑģÐºÐ»Ð°Ð´

0.15

 Cong

0.15

enders

0.15

 Ø£Øº

0.15

.scalablytyped

0.14

uggy

0.14

worthy

0.14

Activations Density 0.258%

phrases related to national interests and political objectives

No Comments

No Known Activations