INDEX

Explanations

instances of discontent or negative emotional expressions

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

ãĥ³ãĥĦ

-0.17

elpers

-0.16

petto

-0.16

adiens

-0.15

lobs

-0.15

inyin

-0.15

ackson

-0.15

å¤§äºº

-0.15

chio

-0.15

etat

-0.15

POSITIVE LOGITS

ep

0.15

0.14

 kali

0.14

ampus

0.14

uj

0.14

 unlimited

0.14

 micro

0.14

ãĥ©ãĤ¯

0.14

ahlen

0.13

gre

0.13

Activations Density 0.004%

instances of discontent or negative emotional expressions

No Comments

No Known Activations