INDEX

Explanations

elements related to typesetting, specifically focusing on characters, letters, symbols, and their representation

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

Tokens related to letters and characters

np_acts-logits-general · gemini-2.0-flash

letters and characters

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

MemoryWarning

-0.46

DrawerToggle

-0.44

>${

-0.44

NavController

-0.43

AddTagHelper

-0.43

')(

-0.43

")(

-0.43

enumii

-0.39

StoryboardSegue

-0.39

 AllMovie

-0.38

POSITIVE LOGITS

 AttributeSet

0.57

 letters

0.57

rungsseite

0.57

 betweenstory

0.54

letters

0.52

 handwriting

0.52

 Administrativna

0.50

 graffiti

0.50

 calligraphy

0.50

alphabet

0.48

Activations Density 0.310%

elements related to typesetting, specifically focusing on characters, letters, symbols, and their representation

Tokens related to letters and characters

letters and characters

No Comments

No Known Activations