INDEX

Explanations

metadata related to software applications and their structure

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

identifier/attribute names

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

ftagPool

-1.03

InputBorder

-0.95

 Normdatei

-0.84

AndEndTag

-0.83

tagHelperRunner

-0.82

enderror

-0.80

 فريبيس

-0.79

:✨

-0.77

 ModelExpression

-0.77

httphttps

-0.75

POSITIVE LOGITS

eaza

0.31

 alternative

0.31

 tambien

0.30

 başar

0.30

ea

0.30

كة

0.29

im

0.29

šč

0.28

im

0.28

<eos>

0.28

Activations Density 0.053%

metadata related to software applications and their structure

identifier/attribute names

No Comments

No Known Activations