INDEX

Explanations

references to types or categories of objects and methods

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

type and classification

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

GEBURTSDATUM

-0.80

 فريبيس

-0.65

 للمعارف

-0.63

 ſind

-0.63

ſelves

-0.60

contentLoaded

-0.59

rbrakk

-0.59

ſelf

-0.57

 queſta

-0.57

POSITIVE LOGITS

 type

0.50

 typu

0.40

 tipo

0.40

 סוג

0.39

 タイプ

0.38

 special

0.36

 Haupts

0.35

 Spezial

0.35

RequestType

0.35

 tipe

0.35

Activations Density 0.040%

references to types or categories of objects and methods

type and classification

No Comments

No Known Activations