INDEX

Explanations

dialogic exchanges and rhetorical questions

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

questions and assertions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

WaitGroup

-0.50

ThemeOverlay

-0.47

 AssemblyCompany

-0.46

NameInMap

-0.46

."]

-0.44

 ligiloj

-0.44

WriteAttribute

-0.42

AddTagHelper

-0.41

__]

-0.41

}));

-0.40

POSITIVE LOGITS

 Administrativna

0.48

 bVar

0.46

不就是

0.44

macos

0.43

ecutable

0.42

 duração

0.42

 bättre

0.41

InstrumentedTest

0.41

 håll

0.41

ませんか

0.41

Activations Density 0.035%

dialogic exchanges and rhetorical questions

questions and assertions

No Comments

No Known Activations