INDEX

Explanations

phrases indicating uncertainty or questioning

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

First-person pronouns after a comma

np_acts-logits-general · gemini-2.0-flash

opinion or personal statement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 30-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_30/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.30.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

 мои

-0.57

 meinen

-0.56

 aking

-0.56

我的

-0.51

 moje

-0.51

my

-0.50

 моих

-0.49

meu

-0.49

 minhas

-0.48

 meus

-0.48

POSITIVE LOGITS

1.73

we

0.75

aš

0.63

我覺得

0.63

 creemos

0.62

 Personally

0.62

0.61

我觉得

0.61

我想

0.59

Personally

0.57

Activations Density 0.977%

phrases indicating uncertainty or questioning

First-person pronouns after a comma

opinion or personal statement

No Comments

No Known Activations