INDEX

Explanations

names and relationships in familial and historical contexts

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

zbollah

-0.16

ç¿Ķ

-0.12

agos

-0.12

 nostalgia

-0.12

HeaderCode

-0.12

 nostalgic

-0.12

\Backend

-0.12

 æ¬

-0.12

è´£

-0.12

á»ĳi

-0.12

POSITIVE LOGITS

 Marriage

0.28

 marriage

0.28

 Married

0.26

 married

0.25

 census

0.24

dau

0.24

 burial

0.23

 death

0.22

 Death

0.22

 Census

0.22

Activations Density 0.182%

names and relationships in familial and historical contexts

No Comments

No Known Activations