INDEX

Explanations

mathematical expressions and programming syntax

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

numbers and symbols

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

AccessorTable

-0.43

 hold

-0.41

 segu

-0.37

Van

-0.36

 signed

-0.35

并将

-0.35

 Performance

-0.35

ftagPool

-0.35

 forgive

-0.35

 passed

-0.35

POSITIVE LOGITS

rungsseite

0.89

0.69

 ddelweddau

0.57

 للمعارف

0.56

SharedDtor

0.56

KommentareTeilen

0.51

 DbSet

0.50

 idéia

0.49

 HasFactory

0.49

 disambiguazione

0.48

Activations Density 0.001%

mathematical expressions and programming syntax

numbers and symbols

No Comments

No Known Activations