INDEX

Explanations

tokens and identifiers related to programming languages and APIs

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

Code-related tokens after capitalized words

np_acts-logits-general · gemini-2.0-flash

technical identifiers followed by separators or keywords

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

EDEFAULT

-0.78

TemporalType

-0.74

ymce

-0.73

Geplaatst

-0.68

DeleteBehavior

-0.63

 SwitchCompat

-0.62

AnimationsModule

-0.62

elemField

-0.62

ChildScrollView

-0.61

featureID

-0.60

POSITIVE LOGITS

 pleaſure

0.76

 reaſon

0.71

 leaſt

0.71

 ſtate

0.70

 purpoſe

0.70

 Diſ

0.65

 ſmall

0.64

 houſe

0.64

 propriety

0.64

 cauſe

0.63

Activations Density 0.738%

tokens and identifiers related to programming languages and APIs

Code-related tokens after capitalized words

technical identifiers followed by separators or keywords

No Comments

No Known Activations