INDEX

Explanations

military ranks and honors

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

military actions and honors

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

хьтан

-0.56

DoubleQuotes

-0.43

Халык

-0.41

 linkovi

-0.41

CppMethod

-0.41

 censiti

-0.39

 pageContext

-0.38

 مشين

-0.38

BASEPATH

-0.38

MessageState

-0.37

POSITIVE LOGITS

 bravely

0.52

 fought

0.52

 valiant

0.50

 courageous

0.50

fought

0.49

ämp

0.47

endorff

0.47

 heroic

0.46

 partici

0.46

 brave

0.46

Activations Density 0.074%

military ranks and honors

military actions and honors

No Comments

No Known Activations