INDEX

Explanations

terms related to devices and their functionalities

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

followed by punctuation or articles

np_acts-logits-general · gemini-2.0-flash

law, crime, and processes

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

webElementXpaths

-0.82

 يتيمه

-0.65

IndentedString

-0.65

 invokingState

-0.60

 jadx

-0.54

Numerade

-0.54

 '\\;'

-0.53

⟬

-0.53

 imagui

-0.52

windowFixed

-0.52

POSITIVE LOGITS

rawtypes

0.35

 stag

0.32

finanz

0.30

เกิด

0.30

 aconte

0.30

 happen

0.30

 حسن

0.29

 Diwali

0.29

 continúas

0.29

 MISS

0.29

Activations Density 1.115%

terms related to devices and their functionalities

followed by punctuation or articles

law, crime, and processes

No Comments

No Known Activations