INDEX

Explanations

copyright-related terms and licensing information

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

<start_of_turn>user queries

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

 вида

-0.33

 bornes

-0.31

 judul

-0.30

 branching

-0.29

 Pseud

-0.29

 бата

-0.29

しき

-0.29

 fill

-0.29

 bakso

-0.29

 embarrassed

-0.29

POSITIVE LOGITS

 kasarigan

0.86

UrlResolution

0.69

OGND

0.68

transQ

0.68

+#+

0.61

outSlope

0.58

httphttps

0.57

 Chwiliwch

0.56

interopRequire

0.55

xase

0.54

Activations Density 0.011%

copyright-related terms and licensing information

<start_of_turn>user queries

No Comments

No Known Activations