INDEX

Explanations

words related to respect and disapproval in various contexts

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

-0.16

to

-0.15

emann

-0.14

andbox

-0.14

¤í

-0.14

rette

-0.14

åĪ°

-0.14

toMatch

-0.14

âĨ

-0.14

alte

-0.14

POSITIVE LOGITS

 widely

0.26

Across

0.22

 across

0.22

 throughout

0.20

 Across

0.20

 widespread

0.20

wid

0.19

 worldwide

0.19

 everywhere

0.19

 among

0.18

Activations Density 0.077%

words related to respect and disapproval in various contexts

No Comments

No Known Activations