INDEX

Explanations

references to items or actions that are divided or handled individually rather than collectively

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

ivan

-0.15

EAR

-0.14

-UA

-0.14

lei

-0.14

Ø¶

-0.14

amar

-0.14

 nÃ¸

-0.14

seau

-0.14

ëłĩ

-0.13

POSITIVE LOGITS

 unrelated

0.17

avra

0.16

 Podesta

0.15

ysa

0.15

æĸŃ

0.14

obble

0.14

imity

0.14

culate

0.14

alte

0.14

unta

0.14

Activations Density 0.024%

references to items or actions that are divided or handled individually rather than collectively

No Comments

No Known Activations