INDEX

Explanations

technical terms and parameters related to data analysis and mathematical concepts

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

cine replay, represented by, history of stroke

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

 surla

-0.69

DoubleQuotes

-0.57

WireFormatLite

-0.53

arakhand

-0.53

enderror

-0.51

ismet

-0.50

󠁣

-0.49

 يتيمه

-0.48

MLLoader

-0.48

Hochspringen

-0.47

POSITIVE LOGITS

Хьажоргаш

0.38

 kijk

0.37

 Offisielt

0.36

pertina

0.36

 בלב

0.35

 arată

0.33

strpos

0.33

 desierto

0.33

intă

0.33

Unters

0.33

Activations Density 0.143%

technical terms and parameters related to data analysis and mathematical concepts

cine replay, represented by, history of stroke

No Comments

No Known Activations