INDEX

Explanations

phrases expressing desire or intention

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

want to

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

-0.61

 HasFactory

-0.58

principalColumn

-0.56

PropertyChanging

-0.56

 AssemblyVersion

-0.56

 виправивши

-0.54

➯

-0.49

tvguidetime

-0.49

ậc

-0.47

Parcelize

-0.47

POSITIVE LOGITS

addColumn

0.45

 yourself

0.44

 Beispiele

0.43

If

0.43

examples

0.42

SequentialGroup

0.42

 examples

0.41

 незавершена

0.40

yourself

0.39

 example

0.38

Activations Density 0.024%

phrases expressing desire or intention

want to

No Comments

No Known Activations