INDEX

Explanations

questions starting with "how."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

uisse

-0.21

ickerView

-0.17

 nettsteder

-0.14

Ð³Ð¾Ð²

-0.14

_OBJC

-0.14

:numel

-0.14

okud

-0.14

RAINT

-0.14

chwitz

-0.14

åĴ¨

-0.14

POSITIVE LOGITS

soever

0.33

 exactly

0.29

/if

0.28

 best

0.27

 they

0.26

itzer

0.26

beit

0.25

 much

0.25

we

0.25

 else

0.23

Activations Density 0.081%

questions starting with "how."

No Comments

No Known Activations