INDEX

Explanations

references to weapons and combat scenarios

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

 Shoot

-0.20

 Shooter

-0.19

Shoot

-0.19

 Guns

-0.18

 guns

-0.18

 shoot

-0.17

 shooting

-0.17

ooter

-0.17

 Shooting

-0.17

gun

-0.16

POSITIVE LOGITS

 rounds

0.27

 shells

0.24

RPG

0.23

 salv

0.23

 stray

0.23

 Rounds

0.22

 tracer

0.20

 Volley

0.20

 blanks

0.20

 volley

0.19

Activations Density 0.095%

references to weapons and combat scenarios

No Comments

No Known Activations