INDEX

Explanations

references to individuals in leadership or authoritative roles, specifically the term "captain."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

 precious

-0.17

utom

-0.14

att

-0.14

 setFrame

-0.14

iert

-0.14

agg

-0.14

away

-0.14

 Inventory

-0.14

æ¢

-0.14

las

-0.14

POSITIVE LOGITS

erville

0.17

wahl

0.17

cy

0.16

ompiler

0.16

 calling

0.15

uzzi

0.15

stvo

0.14

åĳ¼

0.14

oni

0.14

ypo

0.14

Activations Density 0.007%

references to individuals in leadership or authoritative roles, specifically the term "captain."

No Comments

No Known Activations