INDEX

Explanations

imperative verbs and action-oriented phrases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 itemId

-0.52

bcryptjs

-0.50

 MethodInfo

-0.45

byId

-0.43

stdafx

-0.42

deltaTime

-0.42

 cref

-0.42

 Items

-0.41

 Parsons

-0.41

inFile

-0.41

POSITIVE LOGITS

 noDo

0.73

 Educate

0.55

tagHelperRunner

0.54

 biß

0.54

点此举报

0.54

 Take

0.53

atify

0.52

 Simplify

0.51

पया

0.51

fiez

0.50

Activations Density 0.584%

imperative verbs and action-oriented phrases

No Comments

No Known Activations