INDEX

Explanations

terms related to legal charges and criminal activity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 massac

-0.17

æĿĢ

-0.17

 killers

-0.17

 slaughtered

-0.17

 killed

-0.16

 Kills

-0.16

kills

-0.16

 Kill

-0.16

 kills

-0.15

 Killed

-0.15

POSITIVE LOGITS

 punches

0.34

 physical

0.30

 punching

0.30

 punch

0.29

 punched

0.28

 fists

0.27

 violence

0.27

 assault

0.27

 Punch

0.26

Physical

0.25

Activations Density 0.175%

terms related to legal charges and criminal activity

No Comments

No Known Activations

terms related to legal charges and criminal activity

No Comments

No Known Activations