INDEX

Explanations

confirmatory phrases related to evidence and research findings

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

Comes after nouns describing events, processes, or qualities

np_acts-logits-general · gemini-2.0-flash

discovery, studying, characterized, honor

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

hyrchwyd

-0.74

principalTable

-0.62

 ویکی‌پدی

-0.62

oa̍t

-0.61

CppMethod

-0.61

 estekak

-0.60

 وتسجيلات

-0.59

تقاوى

-0.56

 चीज़ों

-0.55

 мәкалә

-0.54

POSITIVE LOGITS

 typelib

0.41

 yourself

0.40

 oneself

0.39

機会

0.35

mentet

0.34

olia

0.33

 devriez

0.32

 rubric

0.32

 crun

0.31

 myself

0.31

Activations Density 1.114%

confirmatory phrases related to evidence and research findings

Comes after nouns describing events, processes, or qualities

discovery, studying, characterized, honor

No Comments

No Known Activations