INDEX

Explanations

the repeated term "ÐŁÑĢ," indicating a focus on a specific identifiable concept or entity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

oin

-0.18

 Laden

-0.17

ey

-0.15

croll

-0.15

stick

-0.14

RenderingContext

-0.14

urger

-0.14

vron

-0.14

ìŀĲìĿ¸

-0.13

POSITIVE LOGITS

.getInteger

0.14

ifie

0.14

aine

0.14

á»¥

0.14

gh

0.14

ober

0.14

apos

0.14

æį

0.13

umer

0.13

alin

0.13

Activations Density 0.003%

the repeated term "ÐŁÑĢ," indicating a focus on a specific identifiable concept or entity

No Comments

No Known Activations