INDEX

Explanations

phrases that express negation or dismissal, particularly using the term "never."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

aises

-0.14

ught

-0.14

enticator

-0.14

sj

-0.13

arty

-0.13

umph

-0.13

Ã³ÅĤ

-0.13

avatars

-0.13

istical

-0.13

ROP

-0.13

POSITIVE LOGITS

mind

0.30

 mind

0.27

winter

0.24

 underestimate

0.23

ending

0.23

land

0.22

 Mind

0.21

Ending

0.21

 trust

0.21

endum

0.20

Activations Density 0.020%

phrases that express negation or dismissal, particularly using the term "never."

No Comments

No Known Activations