INDEX

Explanations

punctuation marks, particularly semicolons and braces

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

code statement terminators

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 entanto

-0.45

 שוליים

-0.44

 diría

-0.44

antaranya

-0.42

 Öffentlichkeit

-0.42

 desmotivaciones

-0.41

 sekal

-0.41

 nadzieję

-0.40

harapkan

-0.40

 arbejde

-0.40

POSITIVE LOGITS

、

0.68

and

0.66

setVerticalGroup

0.61

性和

0.57

 autorytatywna

0.57

 propOrder

0.56

 ffilmiau

0.52

AddHtmlAttribute

0.50

、

0.49

Activations Density 0.031%

punctuation marks, particularly semicolons and braces

code statement terminators

No Comments

No Known Activations