INDEX

Explanations

expressions of dissent and criticism toward authority or proposals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

sounds

-0.15

mare

-0.14

uter

-0.14

etus

-0.13

inand

-0.13

iams

-0.13

 ná»ķi

-0.13

 interess

-0.13

isi

-0.13

ansen

-0.13

POSITIVE LOGITS

 perceived

0.32

 lack

0.28

lack

0.23

 practices

0.22

 decision

0.22

 decisions

0.22

 handling

0.21

 alleged

0.21

way

0.20

 treatment

0.20

Activations Density 0.253%

expressions of dissent and criticism toward authority or proposals

No Comments

No Known Activations