INDEX

Explanations

phrases and references pertaining to conditions or standards of information and objectivity

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

of law

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

ⓧ

-0.53

ThroughAttribute

-0.42

 waer

-0.37

Distribuzione

-0.36

制

-0.36

<h1>

-0.36

able

-0.35

 чем

-0.34

The

-0.34

 förs

-0.34

POSITIVE LOGITS

 getItemCount

0.59

Rohy

0.57

 queſta

0.56

tvguidetime

0.55

__(/*!

0.55

celotti

0.54

 betweenstory

0.54

Демографія

0.54

$_"

0.53

migrationBuilder

0.53

Activations Density 0.045%

phrases and references pertaining to conditions or standards of information and objectivity

of law

No Comments

No Known Activations