INDEX

Explanations

conversational expressions or discourse markers

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

interruption or pause

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

 Мексичка

-0.76

 ویکی‌پدی

-0.73

تقاوى

-0.68

sizeCache

-0.65

featureID

-0.64

 Roskov

-0.63

 Italijani

-0.62

 estekak

-0.62

 Numerade

-0.60

CodeAttribute

-0.59

POSITIVE LOGITS

 Bradley

0.42

 bedenken

0.41

visel

0.41

 pokok

0.40

 another

0.40

Biografie

0.40

sen

0.40

 situation

0.39

dingen

0.39

number

0.39

Activations Density 0.039%

conversational expressions or discourse markers

interruption or pause

No Comments

No Known Activations