INDEX

Explanations

various forms of engagement or interaction in communication

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

First word of a new turn

np_acts-logits-general · gemini-2.0-flash

writing and asking questions

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

Distribuzione

-0.52

 whisky

-0.48

 reggae

-0.47

Referanser

-0.47

Sklici

-0.46

Referencies

-0.45

 inappropriés

-0.45

sidemargin

-0.44

 daff

-0.44

Hozzáférés

-0.44

POSITIVE LOGITS

WireFormatLite

0.57

 betweenstory

0.48

 uLocal

0.47

extAlignment

0.45

UnifiedTopology

0.45

TheReal

0.45

ThemeOverlay

0.42

 AspNetCore

0.42

 Worth

0.40

ItemBackground

0.40

Activations Density 0.950%

various forms of engagement or interaction in communication

First word of a new turn

writing and asking questions

No Comments

No Known Activations