INDEX

Explanations

references to statistical functions or data structures

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

stat and file information

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

الحياه

-0.73

 kasarigan

-0.67

tvguidetime

-0.62

ագրություններ

-0.60

extAlignment

-0.59

новниш

-0.59

 Reverso

-0.58

፩

-0.58

UserScript

-0.58

WebElementEntity

-0.58

POSITIVE LOGITS

stat

0.52

 stat

0.47

FileInfo

0.46

 statute

0.43

 statue

0.41

 statut

0.40

AccessFile

0.38

 sizeof

0.37

fileInfo

0.36

 clín

0.35

Activations Density 0.014%

references to statistical functions or data structures

stat and file information

No Comments

No Known Activations