INDEX

Explanations

terms related to online content and news articles

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

company branding and logos

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

oredCriteria

-0.77

 snippetHide

-0.66

jsxFileName

-0.66

IUrlHelper

-0.65

 HttpNotFound

-0.64

Datuak

-0.63

awaiter

-0.61

 surla

-0.60

Хьажоргаш

-0.60

localctx

-0.60

POSITIVE LOGITS

 signature

0.55

 company

0.44

 logo

0.42

 team

0.42

 Signature

0.42

 employee

0.40

 branding

0.39

 owned

0.38

 staff

0.37

 brand

0.37

Activations Density 0.008%

terms related to online content and news articles

company branding and logos

No Comments

No Known Activations