INDEX

Explanations

terms and phrases related to coercion or compulsion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

roperty

-0.16

UNCH

-0.16

 vast

-0.16

 fisse

-0.16

æ´¥

-0.16

verted

-0.15

Ð¶Ð´

-0.15

aho

-0.15

-webpack

-0.15

eso

-0.14

POSITIVE LOGITS

ful

0.40

fulness

0.30

fully

0.27

-feed

0.26

FUL

0.26

ps

0.25

ive

0.23

-fed

0.23

full

0.22

ythe

0.22

Activations Density 0.047%

terms and phrases related to coercion or compulsion

No Comments

No Known Activations