INDEX

Explanations

references to legal documents or court cases

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

names and addresses

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

 betweenstory

-0.50

 ویکی‌پدی

-0.47

zzleHttp

-0.46

NameInMap

-0.44

endpush

-0.43

jsonwebtoken

-0.42

 snippetHide

-0.41

transQ

-0.40

AdapterFactory

-0.40

 صوتيه

-0.40

POSITIVE LOGITS

 kasarigan

0.60

Ligações

0.44

 bParam

0.43

pieler

0.42

istán

0.41

 Cara

0.41

 оригіналу

0.40

дется

0.40

 propTypes

0.39

Activations Density 0.040%

references to legal documents or court cases

names and addresses

No Comments

No Known Activations