INDEX

Explanations

proper nouns and technical terms

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

writeln, ranos, McMaster, Advanced

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 cherchés

-0.53

Personendaten

-0.50

ValueStyle

-0.50

Aktualisiert

-0.48

 numberWith

-0.46

 Wartawan

-0.44

 NSCoder

-0.43

XtraEditors

-0.43

脚注の使い方

-0.42

PCell

-0.40

POSITIVE LOGITS

 calldata

0.76

AEW

0.67

ranos

0.66

opardy

0.65

edb

0.63

UVWXYZ

0.62

周泽楷

0.54

writeln

0.53

gerton

0.52

{?}

0.49

Activations Density 0.004%

proper nouns and technical terms

writeln, ranos, McMaster, Advanced

No Comments

No Known Activations

proper nouns and technical terms

writeln, ranos, McMaster, Advanced

No Comments

No Known Activations