INDEX

Explanations

references to Hall of Fame inductions or memberships

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

Animator

-0.15

 chall

-0.15

æŁ³

-0.15

Ð¸Ð½Ð¾Ðº

-0.14

hiro

-0.14

 prac

-0.14

vertime

-0.14

odal

-0.13

Lic

-0.13

rou

-0.13

POSITIVE LOGITS

 induction

0.51

 indu

0.42

Ind

0.41

 Hall

0.38

Hall

0.33

ducted

0.33

 hall

0.32

Ind

0.32

duct

0.32

-ind

0.29

Activations Density 0.043%

references to Hall of Fame inductions or memberships

No Comments

No Known Activations