INDEX

Explanations

words related to notable or infamous characters, specifically their names or titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

worth

-0.17

aine

-0.16

wise

-0.16

--

-0.16

witter

-0.15

 Rapid

-0.14

pn

-0.14

etwork

-0.14

ulo

-0.14

 Strike

-0.14

POSITIVE LOGITS

obao

0.17

Ã©o

0.17

urette

0.16

 Griff

0.15

named

0.14

/MIT

0.14

SFML

0.14

deen

0.14

ifu

0.14

ÃŃÅ¾

0.14

Activations Density 0.001%

words related to notable or infamous characters, specifically their names or titles

No Comments

No Known Activations