INDEX

Explanations

concepts related to authority or dominance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

moz

-0.16

wards

-0.15

Ale

-0.14

skill

-0.14

mos

-0.14

Unused

-0.14

ynam

-0.13

 Holt

-0.13

zi

-0.13

POSITIVE LOGITS

 supreme

0.36

 Supreme

0.29

 reign

0.23

ited

0.20

 over

0.20

ite

0.18

 terror

0.18

ITE

0.17

 monarch

0.17

iting

0.17

Activations Density 0.026%

concepts related to authority or dominance

No Comments

No Known Activations