INDEX

Explanations

the word "There" indicating presence or existence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

chia

-0.17

sth

-0.16

ttp

-0.15

»

-0.15

venes

-0.15

 here

-0.15

atego

-0.14

reek

-0.14

repid

-0.14

 Antar

-0.14

POSITIVE LOGITS

 they

0.17

igh

0.15

after

0.15

lasting

0.14

 dort

0.14

UniqueId

0.14

plit

0.14

alone

0.13

upon

0.13

ADER

0.13

Activations Density 0.052%

the word "There" indicating presence or existence

No Comments

No Known Activations