INDEX

Explanations

references to military conflicts or confrontations

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

 hoÃłn

-0.15

éĳ

-0.14

ilon

-0.14

CTION

-0.14

ction

-0.14

alendar

-0.13

 maneuver

-0.13

OLON

-0.13

 Slash

-0.13

BY

-0.13

POSITIVE LOGITS

åłĤ

0.18

èĥĨ

0.16

Falsy

0.16

iver

0.15

loit

0.14

Î²Î¹

0.14

_WP

0.14

 palindrome

0.13

 Lessons

0.13

\core

0.13

Activations Density 0.001%

references to military conflicts or confrontations

No Comments

No Known Activations