INDEX

Explanations

descriptors related to violent actions and their consequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

 creep

-0.17

 overnight

-0.17

beck

-0.16

 knots

-0.14

 Finger

-0.14

 Overflow

-0.14

 Battles

-0.14

 handjob

-0.14

 creeping

-0.14

ahn

-0.14

POSITIVE LOGITS

 impact

0.30

Impact

0.28

impact

0.27

 Impact

0.25

 impacts

0.24

 landing

0.24

landing

0.21

 impacting

0.21

 Landing

0.20

 rico

0.20

Activations Density 0.380%

descriptors related to violent actions and their consequences

No Comments

No Known Activations