INDEX

Explanations

conversational expressions and emotional responses

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

questions and reactions

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

uxxxx

-0.52

Gambas

-0.43

 espor

-0.42

 Италијани

-0.42

 monitoring

-0.41

ChildScrollView

-0.41

LETIN

-0.41

Focusing

-0.41

Monitoring

-0.40

 Monitoring

-0.40

POSITIVE LOGITS

 समीक्षाओं

0.48

 reaction

0.48

 react

0.44

MENAFN

0.43

rawDesc

0.42

 reactions

0.40

 Reaktion

0.39

 disbelief

0.38

UnitTesting

0.36

 resourceCulture

0.36

Activations Density 0.032%

conversational expressions and emotional responses

questions and reactions

No Comments

No Known Activations