INDEX

Explanations

phrases that express curiosity or contemplation

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

zman

-0.17

.scalablytyped

-0.17

inerary

-0.16

ussen

-0.16

rimp

-0.15

side

-0.15

ãĥ¼ãĥĢ

-0.15

idl

-0.14

ÅĽÄĩ

-0.14

lassian

-0.14

POSITIVE LOGITS

 whether

0.20

ous

0.17

ously

0.17

whether

0.16

om

0.16

how

0.16

why

0.15

 Jacobs

0.15

lust

0.15

ingly

0.14

Activations Density 0.023%

phrases that express curiosity or contemplation

No Known Activations