INDEX

Explanations

references to organizations, networks, or entities involved in health and wellness

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

company locations and contact info

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

]")]

-0.88

 فريبيس

-0.84

AndEndTag

-0.84

 autorytatywna

-0.83

ValueStyle

-0.82

 Taktlose

-0.82

 Chwiliwch

-0.81



-0.79

__":

-0.79

цездатний

-0.79

POSITIVE LOGITS

<eos>

0.43

0.41

0.35

por

0.35

//

0.35

 bety

0.34

TM

0.33

 Adamson

0.33

ขาว

0.32

Activations Density 0.010%

references to organizations, networks, or entities involved in health and wellness

company locations and contact info

No Comments

No Known Activations