INDEX

Explanations

phrases expressing emotional outrage or conflict

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

Followed by punctuation/exclamation

np_acts-logits-general · gemini-2.0-flash

aggressive punctuation and commands

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

)?;

-0.54

 вроде

-0.52

"}";

-0.50

çası

-0.49

<>());

-0.47

]--;

-0.46

HEADERS

-0.46

ㅜ

-0.46

"}")

-0.46

HostException

-0.46

POSITIVE LOGITS

 autorytatywna

0.81

🖕

0.81

 ***!

0.72

Shut

0.62

TypedDataSet

0.60

ThroughAttribute

0.59

COUVER

0.59

المناصب

0.57

StructEnd

0.56

MessageOf

0.56

Activations Density 0.140%

phrases expressing emotional outrage or conflict

Followed by punctuation/exclamation

aggressive punctuation and commands

No Comments

No Known Activations