INDEX

Explanations

references to military and defense-related topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

EMS

-0.14

STS

-0.13

 Telescope

-0.13

 Emanuel

-0.13

bilt

-0.13

wing

-0.13

 microscope

-0.13

dispatch

-0.13

pud

-0.13

.Dispatcher

-0.13

POSITIVE LOGITS

 mine

0.46

 mines

0.43

 Mine

0.40

mine

0.39

 Mines

0.37

Mine

0.35

 mined

0.30

 Clearance

0.29

 land

0.29

 clearance

0.29

Activations Density 0.009%

references to military and defense-related topics

No Comments

No Known Activations