INDEX

Explanations

instances of the word "remember" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

vet

-0.16

ifa

-0.15

.sponge

-0.15

ilon

-0.14

Ã¤t

-0.14

ader

-0.14

abilecek

-0.14

reve

-0.13

oka

-0.13

isse

-0.13

POSITIVE LOGITS

 fond

0.31

 distinctly

0.27

 vivid

0.26

 distinct

0.25

how

0.25

distinct

0.23

 DISTINCT

0.21

 Fond

0.21

 clearly

0.21

 hearing

0.20

Activations Density 0.033%

instances of the word "remember" and its variations

No Comments

No Known Activations