INDEX

Explanations

programming constructs and data structures, particularly in Java syntax

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

code structure and keywords

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

 Italijani

-0.47

 Thirdly

-0.37

 Administrativna

-0.35

Geplaatst

-0.35

 Secondly

-0.35

GHIJKLM

-0.34

存于互联网档案馆

-0.34



-0.34

 يتيمه

-0.33

 Roskov

-0.31

POSITIVE LOGITS

OGND

0.67

DockStyle

0.59

principalTable

0.53

 organisms

0.51

Personensuche

0.50

 armchair

0.49

 bailarina

0.49

TestingModule

0.49

 dancer

0.48

 suprême

0.47

Activations Density 0.011%

programming constructs and data structures, particularly in Java syntax

code structure and keywords

No Comments

No Known Activations