INDEX

Explanations

concepts related to narration and storytelling in a structural sense

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

computing model

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

 sticker

-0.41

SourceChecksum

-0.40

 Sticker

-0.39

cracker

-0.39

WriteAttribute

-0.38

 grade

-0.38

 sele

-0.38

 Sheet

-0.37

ägg

-0.37

nok

-0.36

POSITIVE LOGITS

 nahilalakip

0.64

 surla

0.50

 ostavi

0.48

 Accesat

0.46

 تكبرها

0.46

 informée

0.46

 setMessage

0.45



0.44

 otomatig

0.43

delwed

0.43

Activations Density 0.189%

concepts related to narration and storytelling in a structural sense

computing model

No Comments

No Known Activations