INDEX

Explanations

the presence of various linking verbs and their associations with other words

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

preceding "are," "were," or "is."

np_acts-logits-general · gemini-2.0-flash

is/are/were followed by descriptor

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

styleUrls

-0.39

borderFill

-0.36

ugno

-0.36

ammer

-0.36

 excru

-0.35

 Jika

-0.35

Tikang

-0.35

srand

-0.34

Fanny

-0.34

utilisons

-0.34

POSITIVE LOGITS

AccessorTable

0.54

TextAppearance

0.53

 nakalista

0.52

 utafitiHapana

0.50

 BoxDecoration

0.49

DebuggerNonUser

0.49

rzez

0.47

QMetaType

0.46

 TextAppearance

0.45

 resultCode

0.44

Activations Density 0.341%

the presence of various linking verbs and their associations with other words

preceding "are," "were," or "is."

is/are/were followed by descriptor

No Comments

No Known Activations