INDEX

Explanations

references to specifications and characteristics of products or experiments

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

followed by punctuation or prepositions

np_acts-logits-general · gemini-2.0-flash

parenthesized numbers, units, or labels

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

 Administrativna

-0.44

 Audiodateien

-0.43

 istruzioni

-0.36

 EventArgs

-0.35

AnchorStyles

-0.33

差

-0.33

 realidad

-0.33

 experiencia

-0.32

 navideñas

-0.32

hdessä

-0.32

POSITIVE LOGITS

 Chwiliwch

0.71

 AssemblyTitle

0.68

تقاوى

0.66

GIVEREF

0.64

ніципалі

0.62

 Signalez

0.60

rungsseite

0.59

 незавершена

0.57

 ligiloj

0.57

@[+][

0.57

Activations Density 1.013%

references to specifications and characteristics of products or experiments

followed by punctuation or prepositions

parenthesized numbers, units, or labels

No Comments

No Known Activations