INDEX

Explanations

programming terms and concepts related to file and directory management

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

resolve, search, directories

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

 ModelExpression

-0.59

UnusedPrivate

-0.59

uxxxx

-0.50

 defaultstate

-0.48

 jadx

-0.47

EnableWeb

-0.45

 חיצוניים

-0.45

 theat

-0.44

éez

-0.43

ɵɵelementStart

-0.43

POSITIVE LOGITS

SharedCtor

0.55

 pencarian

0.50

ComponentScan

0.50

resolve

0.47

يكب

0.47

PreferredItem

0.47

 obuv

0.45

prefixer

0.43

 equipaje

0.42

Searching

0.41

Activations Density 0.002%

programming terms and concepts related to file and directory management

resolve, search, directories

No Comments

No Known Activations