INDEX

Explanations

conversational elements related to personal experiences or problems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 pinulongan

-0.51

hyrchwyd

-0.47

IUrlHelper

-0.46

鰭

-0.46

 Мексичка

-0.44

 âmes

-0.44

MethodManager

-0.43

匿名使用者

-0.43

Hentet

-0.43

ագրություններ

-0.43

POSITIVE LOGITS

IVEREF

0.42

 ComVisible

0.40

endpush

0.40

Exclu

0.39

 trag

0.39

 sekali

0.39

PYX

0.39

 Paglinawan

0.38

phat

0.37

glow

0.37

Activations Density 0.328%

conversational elements related to personal experiences or problems

No Comments

No Known Activations