INDEX

Explanations

titles or labels of scientific papers or studies

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

objs

-0.13

 volunteers

-0.13

ic

-0.13

 pont

-0.13

fone

-0.13

.aspx

-0.13

 Tyson

-0.13

çĭĲ

-0.13

 Commerce

-0.13

Hog

-0.12

POSITIVE LOGITS

ahat

0.16

ardo

0.15

 TRACE

0.14

Sle

0.14

Ø¡

0.14

PIO

0.14

\Backend

0.14

uling

0.14

ume

0.14

lero

0.13

Activations Density 0.028%

titles or labels of scientific papers or studies

No Comments

No Known Activations