INDEX

Explanations

punctuation marks and whitespace

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

technical or measurement data

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

 kaarangay

-1.00

rungsseite

-0.89

⟬

-0.87

 betweenstory

-0.85

RegressionTest

-0.83

transQ

-0.83

[@BOS@]

-0.82

<unused41>

-0.82

<unused28>

-0.82

<unused14>

-0.82

POSITIVE LOGITS

0.32

0.28

 Bridge

0.27

ImageName

0.24

mode

0.23

..

0.23

 autorytatywna

0.23

.,

0.23

 Google

0.23

MODE

0.23

Activations Density 0.000%

punctuation marks and whitespace

technical or measurement data

No Comments

No Known Activations