INDEX

Explanations

references to locations and specific numeric data

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

years and specific dates

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

 utafitiHapana

-0.36

)))));

-0.35

jugend

-0.33

횟

-0.33

endgroup

-0.33

 aktuellen

-0.33

 replies

-0.32

/******/

-0.32

du

-0.32

 copa

-0.32

POSITIVE LOGITS

writeFieldEnd

0.65

 propOrder

0.64

 initComponents

0.62

 informée

0.56

parsedMessage

0.53

 Италијани

0.52

 للمعارف

0.50

 defaultstate

0.49

 ویکی‌پدی

0.48

èdia

0.48

Activations Density 0.095%

references to locations and specific numeric data

years and specific dates

No Comments

No Known Activations