INDEX

Explanations

complex string patterns that may represent data structures or indices

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

Hexadecimal byte values

np_acts-logits-general · gemini-2.0-flash

tokens followed by a comma

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

="@+

-0.66

ManagedBean

-0.61

 المعيارى

-0.61

migrationBuilder

-0.59

 Приступљено

-0.58

ActionCreators

-0.56

UrlResolution

-0.54

 onCreateView

-0.53

ACHUSET

-0.53

rungsseite

-0.53

POSITIVE LOGITS

 Turquía

0.42

 myſelf

0.41

 Comunicação

0.41

 niyang

0.40

 izquier

0.40

 enfans

0.38

 Kerja

0.37

 Theſe

0.37

 Beſ

0.37

 Gestão

0.36

Activations Density 1.308%

complex string patterns that may represent data structures or indices

Hexadecimal byte values

tokens followed by a comma

No Comments

No Known Activations