INDEX

Explanations

references to individual names and their roles or actions

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

Reporting speech

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



-0.55

cani

-0.51

 CascadeType

-0.46

 ویکی‌آمباردا

-0.46

basicConfig

-0.45

Utf

-0.45

хьтан

-0.44

まま

-0.44

 nonUne

-0.44

ViewChild

-0.43

POSITIVE LOGITS

RenderAtEndOf

0.39

 menjawab

0.37

setVerticalGroup

0.35

 defesa

0.34

 praises

0.33

ViewFeatures

0.33

 المعيارى

0.33

 aikaa

0.32

antwortung

0.32

 Inscrivez

0.32

Activations Density 0.015%

references to individual names and their roles or actions

Reporting speech

No Comments

No Known Activations