INDEX

Explanations

numerical rankings or results related to competitive events

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

1. numbered names

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

WriteLiteral

-0.67

tagHelperRunner

-0.49

transQ

-0.48

webElementXpaths

-0.48

IUrlHelper

-0.47

 banking

-0.46

ыгана

-0.46

 Banking

-0.45

 attitude

-0.45

crisy

-0.44

POSITIVE LOGITS

SerializedName

0.54

 nahilalakip

0.52

awtextra

0.52

StoryboardSegue

0.44

ReusableCell

0.42

 newBuilder

0.41

 مشين

0.40

 GenerationType

0.39

CloseOperation

0.39

 doInBackground

0.37

Activations Density 0.024%

numerical rankings or results related to competitive events

1. numbered names

No Comments

No Known Activations