INDEX

Explanations

programming language annotations and constructs

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

special sequence markers

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

EnableWeb

-0.50

ValueGeneration

-0.47

çade

-0.45

masta

-0.43

pert

-0.43

 Oaks

-0.42

 BoxDecoration

-0.41

oaks

-0.40

lito

-0.40

Neg

-0.40

POSITIVE LOGITS

databind

0.78

gdx

0.57

Diweddarwch

0.56

ectl

0.55

MockMvc

0.54

 المعيارى

0.54

CollectionView

0.54

 Muhammadu

0.54

kubectl

0.52

 XCTest

0.52

Activations Density 0.000%

programming language annotations and constructs

special sequence markers

No Comments

No Known Activations