INDEX

Explanations

words and phrases related to provocation and provocative speech

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

utex

-0.14

mad

-0.14

erez

-0.14

ern

-0.14

Rip

-0.14

optera

-0.14

ule

-0.13

çµ¡

-0.13

ensch

-0.13

Ã¤

-0.13

POSITIVE LOGITS

Å¾en

0.18

/assert

0.17

eyin

0.16

CHASE

0.16

Äįin

0.15

zÄħd

0.15

lint

0.14

nutÃŃm

0.14

ãĥ©ãĥĥãĤ¯

0.14

nÃŃk

0.14

Activations Density 0.006%

words and phrases related to provocation and provocative speech

No Comments

No Known Activations