INDEX

Explanations

phrases and words related to personal identity and belonging

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

family and home-related words

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

Rüyada

-0.53

tvguidetime

-0.52

bootstrapcdn

-0.47

ArgumentParser

-0.47

 تانيه

-0.45

PerformLayout

-0.45

BorderFactory

-0.44

StoryboardSegue

-0.44

EnumMember

-0.44

 Signalez

-0.43

POSITIVE LOGITS

 plantel

0.40

 almoço

0.38

épendance

0.37

 générations

0.37

 rafraî

0.36

 bâtiments

0.36

 chaleco

0.36

 mediodía

0.36

olverine

0.35

 sommeil

0.35

Activations Density 0.036%

phrases and words related to personal identity and belonging

family and home-related words

No Comments

No Known Activations