INDEX

Explanations

mathematical expressions and symbols

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

roots and powers

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

 Administrativna

-0.66

 utafitiHapana

-0.65

 initComponents

-0.63

 useAppContext

-0.59

 Italijanski

-0.57

 defaultstate

-0.56

AutoScaleMode

-0.51

PAS

-0.47

 Liability

-0.47

 Савезне

-0.47

POSITIVE LOGITS

∛

0.49

Diweddarwch

0.42

 revolución

0.40

 competencia

0.38

pexpr

0.38

 révolution

0.37

Tikang

0.36

 revolution

0.35

FontOfSize

0.35

=$("#

0.35

Activations Density 0.055%

mathematical expressions and symbols

roots and powers

No Comments

No Known Activations