INDEX

Explanations

terms related to medical or scientific analysis and measurements

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

except, which

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

 للاسماء

-0.58

AnchorStyles

-0.51

Życiorys

-0.49

Hozzáférés

-0.47

JspWriter

-0.47

 BoxDecoration

-0.44

unhofer

-0.44

scriptcase

-0.41

fira

-0.40

CharSequence

-0.40

POSITIVE LOGITS

 виправивши

0.58

OGND

0.57

 Occasionally

0.42

Occasionally

0.42

 occasionally

0.40

nezeu

0.39

 tertentu

0.38

偶尔

0.37

әһ

0.37

 Zink

0.36

Activations Density 0.086%

terms related to medical or scientific analysis and measurements

except, which

No Comments

No Known Activations