INDEX

Explanations

keywords and identifiers related to programming and class structures

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

class derivation

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

/*

-0.47

 vertrouwen

-0.46

Демографія

-0.46

 mijne

-0.45

envolvimento

-0.45

 kohdetta

-0.44

 perasaan

-0.44

 gesprek

-0.44

 parceiros

-0.43

ardless

-0.43

POSITIVE LOGITS

الحياه

0.53

PreAuthorize

0.41

 insuffisamment

0.41

WithIOException

0.39

 autorytatywna

0.39

OrBuilder

0.38

 subclass

0.38

MLLoader

0.38

dafx

0.37

PerformLayout

0.37

Activations Density 0.058%

keywords and identifiers related to programming and class structures

class derivation

No Comments

No Known Activations