INDEX

Explanations

phrases related to questions, corrections, and addressing issues or concerns

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

any

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

뀜

-0.60

parsedMessage

-0.60

multer

-0.52

/**

-0.50

 centaines

-0.47

ioutil

-0.47

EnableWeb

-0.47

rrggbb

-0.46

ollectionView

-0.45

Somebody

-0.45

POSITIVE LOGITS

any

2.77

 cualquier

2.00

Any

1.99

Any

1.98

任何

1.92

 qualquer

1.87

any

1.85

 любые

1.82

 quaisquer

1.78

 любых

1.70

Activations Density 0.399%

phrases related to questions, corrections, and addressing issues or concerns

any

No Comments

No Known Activations