INDEX

Explanations

references to locations and institutions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

iju

-0.17

arella

-0.16

gay

-0.16

.scalablytyped

-0.15

essa

-0.15

RT

-0.15

iani

-0.15

oust

-0.15

åĴ

-0.14

 Ansi

-0.14

POSITIVE LOGITS

íĮĲ

0.16

/-

0.14

 briefly

0.14

ä¾Ľ

0.14

nonnull

0.13

 ACTIVE

0.13

æĿ¡

0.13

 active

0.13

wee

0.13

Activations Density 0.026%

references to locations and institutions

No Comments

No Known Activations