INDEX

Explanations

numerical references and identifiers related to locations or properties

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

 Area

-0.16

 beaches

-0.15

orgot

-0.15

dur

-0.15

amework

-0.14

arov

-0.14

telegram

-0.14

strup

-0.14

 Region

-0.13

Kaf

-0.13

POSITIVE LOGITS

Ø§Ø¨Ø§ÙĨ

0.16

Į¨

0.15

 unlucky

0.15

 podium

0.14

_CLAMP

0.14

aus

0.14

kenin

0.14

portun

0.14

è¡Ĺ

0.13

Î±Î½Î¿Ïħ

0.13

Activations Density 0.100%

numerical references and identifiers related to locations or properties

No Comments

No Known Activations