INDEX

Explanations

intensifiers that emphasize strong emotional or qualitative states

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

oline

-0.18

pic

-0.16

agn

-0.15

ëŁī

-0.15

naments

-0.15

agen

-0.15

ãĥĨãĥ«

-0.15

owa

-0.15

rypt

-0.14

POSITIVE LOGITS

ening

0.32

ened

0.26

thro

0.20

ender

0.17

(er

0.17

ablish

0.17

 deep

0.17

ocre

0.16

deep

0.16

ens

0.15

Activations Density 0.022%

intensifiers that emphasize strong emotional or qualitative states

No Comments

No Known Activations