INDEX

Explanations

phrases indicating moral judgment or hypocrisy in political discourse

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

¼

-0.14

 spyOn

-0.14

 sáºµn

-0.14

oug

-0.14

WithURL

-0.14

Ð½ÐµÐ½

-0.14

 grÃ¢ce

-0.14

lia

-0.14

³ç´°

-0.14

gratis

-0.13

POSITIVE LOGITS

 border

0.23

 borders

0.18

 unless

0.18

border

0.18

 beyond

0.18

-border

0.18

 behavior

0.18

 considering

0.18

 attempted

0.17

 attempt

0.16

Activations Density 0.257%

phrases indicating moral judgment or hypocrisy in political discourse

No Comments

No Known Activations