INDEX

Explanations

phrases containing the word "new" connected to various contexts, particularly in discussions of construction or renovation

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

new car or building

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

 Popov

-0.40

langkah

-0.37

 accept

-0.36

 desideri

-0.34

 accepted

-0.34

 acepta

-0.33

 trovi

-0.32

是一种

-0.32

 initially

-0.31

 नि

-0.31

POSITIVE LOGITS

MessageTagHelper

0.72

脚注の使い方

0.63

posedge

0.63

 hObject

0.60

Personensuche

0.60

 resourceCulture

0.59

 ProtoMessage

0.57

FieldBuilder

0.57

 Reverso

0.56

 sumpay

0.56

Activations Density 0.018%

phrases containing the word "new" connected to various contexts, particularly in discussions of construction or renovation

new car or building

No Comments

No Known Activations