INDEX

Explanations

terms related to titles, roles, or positions, particularly within a leadership context

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

offensive, fundamental, negative, solid

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

exitRule

-0.44

ThroughAttribute

-0.41

 Administrativna

-0.40

contentLoaded

-0.40

IsContent

-0.39

viders

-0.38

исленность

-0.38

Экономи

-0.37

 nød

-0.37

 censiti

-0.36

POSITIVE LOGITS

:✨

0.52

 سكانية

0.45

extAlignment

0.43

 MainAxisSize

0.43

gnore

0.42

derry

0.41

 Gait

0.39

乓

0.38

Myth

0.38

endcsname

0.36

Activations Density 0.451%

terms related to titles, roles, or positions, particularly within a leadership context

offensive, fundamental, negative, solid

No Comments

No Known Activations