INDEX

Explanations

punctuation marks

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

user input context

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

 betweenstory

-1.02

RegressionTest

-0.89

UserScript

-0.84

uxxxx

-0.78

transQ

-0.71

ݮ

-0.65

msgTypes

-0.65

 ostavi

-0.64

évaluateur

-0.63

 Administrativna

-0.62

POSITIVE LOGITS

findpost

0.43

 rainha

0.38

 antidesliz

0.31

 diário

0.31

 embarazada

0.30

 Pendidikan

0.30

 lluvias

0.30

 pengetahuan

0.29

 táctil

0.29

 productividad

0.29

Activations Density 0.001%

punctuation marks

user input context

No Comments

No Known Activations