INDEX

Explanations

connections or patterns related to scientific or technical processes

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

HTML tags and other code snippets

np_acts-logits-general · gemini-2.0-flash

in clauses or phrases

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

Personendaten

-0.65

 Dostupné

-0.64

webElementGuid

-0.55

 Arki

-0.54

ographiques

-0.51

RectangleBorder

-0.50

 defaultstate

-0.50

 Madame

-0.48

PhysRev

-0.48

 bezeichneter

-0.47

POSITIVE LOGITS

\{\\

0.79

0.64

<h4>

0.60

ⓘ

0.58

<h6>

0.55

<table>

0.55

<h5>

0.55

0.54

<td>

0.54



0.54

Activations Density 0.079%

connections or patterns related to scientific or technical processes

HTML tags and other code snippets

in clauses or phrases

No Comments

No Known Activations