INDEX

Explanations

terms related to software components and services in a programming context

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

directive communication

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

 CONTIN

-0.36

 this

-0.33

zahl

-0.33

diğim

-0.32

 lleno

-0.32

odw

-0.31

 Schae

-0.31

 refuer

-0.31

 continuidad

-0.31

orique

-0.30

POSITIVE LOGITS

 StatelessWidget

0.84

 OnInit

0.81

NgModule

0.80

SourceChecksum

0.65

OGND

0.65

 AppComponent

0.64

 createState

0.62

nodoc

0.61

FormTagHelper

0.61

medriver

0.60

Activations Density 0.001%

terms related to software components and services in a programming context

directive communication

No Comments

No Known Activations