INDEX

Explanations

modal verbs indicating speculation or prediction

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

romo

-0.17

utex

-0.17

ozor

-0.15

åĿĬ

-0.15

.ManyToMany

-0.15

arend

-0.15

.mapbox

-0.15

 mÃ¬nh

-0.15

ìķ¼

-0.15

Ã½Å¡

-0.15

POSITIVE LOGITS

 themselves

0.23

 probably

0.20

 likely

0.19

 tell

0.18

arg

0.18

arg

0.18

cr

0.18

 telling

0.18

 modest

0.17

Activations Density 0.117%

modal verbs indicating speculation or prediction

No Comments

No Known Activations