INDEX

Explanations

dialogues and interactions between characters

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

states of emotion or reaction

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

 kasarigan

-0.65

IndentedString

-0.59

tvguidetime

-0.59

таратура

-0.58

PMailer

-0.56

 GenerationType

-0.55

 المعيارى

-0.50



-0.49

 nakalista

-0.49

IntoConstraints

-0.48

POSITIVE LOGITS

Pfe

0.39

 envy

0.38

XmlType

0.37

 eyes

0.36

 loudly

0.36

Quiénes

0.35

cupertino

0.35

 loud

0.35

 labels

0.34

anek

0.34

Activations Density 0.024%

dialogues and interactions between characters

states of emotion or reaction

No Comments

No Known Activations