INDEX

Explanations

programming constructs related to class and interface definitions

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

class and interface definitions

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

 utafitiHapana

-0.66

 Chwiliwch

-0.65

 Normdatei

-0.62

 ocasión

-0.52

 ModelExpression

-0.52

 怎样

-0.51

 ocupación

-0.49

UnusedPrivate

-0.48

 BoxFit

-0.47

PreExecute

-0.47

POSITIVE LOGITS

 createSlice

0.36

хьтан

0.35

endmodule

0.34

Parcelize

0.33

IVEREF

0.32

 Doble

0.32

 Alber

0.31

 Eben

0.31

zeigen

0.31

انتهای

0.31

Activations Density 0.005%

programming constructs related to class and interface definitions

class and interface definitions

No Comments

No Known Activations