INDEX

Explanations

references to gymnastics

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

ypse

-0.17

paced

-0.17

itez

-0.17

ARING

-0.16

pole

-0.16

itaire

-0.16

poke

-0.16

igner

-0.16

 kterÃ©ho

-0.15

uels

-0.15

POSITIVE LOGITS

astics

0.42

nas

0.37

astic

0.32

ast

0.28

asts

0.27

kh

0.25

asi

0.24

nos

0.23

ternet

0.23

astically

0.22

Activations Density 0.007%

references to gymnastics

No Comments

No Known Activations