INDEX

Explanations

proper nouns and names related to people and organizations

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

titles and names

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

 bowiem

-0.46

 ModelExpression

-0.43

ValueGeneration

-0.42

ReusableCell

-0.42

 onOptions

-0.42

ptăm

-0.42

 atve

-0.41

 alugar

-0.41

 avoient

-0.40

 echip

-0.40

POSITIVE LOGITS

Хьажоргаш

0.50

 utafitiHapana

0.47

trab

0.45

IVEREF

0.44

لينكات

0.42

oa̍t

0.40

 propOrder

0.40

StateToProps

0.39

 dialog

0.38

 Langer

0.37

Activations Density 0.106%

proper nouns and names related to people and organizations

titles and names

No Comments

No Known Activations