INDEX

Explanations

references to military training and operations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

å¼

-0.15

ÑĤÑĢÐ¾Ð½

-0.15

 cannons

-0.15

 manoe

-0.14

çĸ

-0.14

éĲµ

-0.14

 Allison

-0.13

 maneuvers

-0.13

yp

-0.13

Ø¬Ø§Ø±

-0.13

POSITIVE LOGITS

 missions

0.18

 doctr

0.18

 litt

0.17

ogne

0.16

 credible

0.16

 Synthetic

0.15

 doctrine

0.15

 theatre

0.14

naissance

0.14

 conducts

0.14

Activations Density 0.127%

references to military training and operations

No Comments

No Known Activations