INDEX

Explanations

references related to programming resources and documentation

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

academic citations like ppat and pone

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

MessageTagHelper

-0.52

 <<<<<<<<<<<<<<

-0.47

 bílá

-0.47

ChrTalk

-0.45

 HttpNotFound

-0.45

ngOnInit

-0.41

 otomatig

-0.40

 demografica

-0.38

 Escolar

-0.37

 Bewußt

-0.37

POSITIVE LOGITS

ecco

0.51

 tartalomajánló

0.50

 Roskov

0.49

Derbyniad

0.49

Tikang

0.49

ACHUSET

0.48

 الحره

0.48

󠁴

0.47

 Réponses

0.47

uq

0.45

Activations Density 0.016%

references related to programming resources and documentation

academic citations like ppat and pone

No Comments

No Known Activations